feat: wire up paged decode CUDA kernel to Python extension

- Add attn_paged_decode wrapper in ops.py with gather fallback
- Register kernel in loader.py and export from __init__.py
- Extract test_utils.cuh shared by all attention unit tests
- Rename attn_paged_vs_contiguous.cu to attn_paged_decode_test.cu
- Refactor decode/prefill tests to use common bf16 helpers and cpu ref
- Fix k_cache dim check in attn_paged_decode.cu

.gitignore

@@ -5,8 +5,16 @@
 !*/
 
 # Allow specific file types and root files
-!*.py
-!*.sh
+!astrai/**/*.py
+!scripts/**/*.py
+!tests/**/*.py
+!csrc/**/*.py
+
+!csrc/**/*.cu
+!csrc/**/*.h
+!csrc/**/*.cuh
+
+!scripts/**/*.sh
 
 # Allow GitHub files
 !/.github/**
@@ -21,3 +29,8 @@
 !/LICENSE
 !/pyproject.toml
 !/README.md
+# Allow extension modules (only source .py)
+!/astrai/extension/**/*.py
+
+# Allow build files
+!/setup.py

CONTRIBUTING.md

@@ -5,7 +5,7 @@ Thank you for your interest in contributing! This document provides step-by-step
 ## Quick Start
 
 ```bash
-git clone https://github.com/your-username/AstrAI.git
+git clone https://github.com/ViperEkura/AstrAI.git
 cd AstrAI
 pip install -e ".[dev]"     # install with dev dependencies (pytest, ruff)
 ```

Dockerfile

@@ -23,7 +23,7 @@ COPY astrai/ ./astrai/
 COPY pyproject.toml .
 RUN pip install --no-cache-dir --upgrade pip \
     && pip install --no-cache-dir . \
-    --extra-index-url https://download.pytorch.org/whl/cu126
+    --extra-index-url https://download.pytorch.org/whl/cu128
 
 # Production stage
 FROM ubuntu:24.04 AS production

README.md

@@ -9,9 +9,9 @@
 <div align="center">
   <img src="https://img.shields.io/badge/python-3.12+-blue.svg" alt="python">
   <img src="https://img.shields.io/badge/license-GPL--3.0-blue.svg" alt="license">
-  <img src="https://img.shields.io/github/v/release/ViperEkura/AstrAI?color=76bad9" alt="release">
-  <img src="https://img.shields.io/badge/dynamic/json?url=https%3A%2F%2Fapi.github.com%2Frepos%2FViperEkura%2FAstrAI&query=%24.stargazers_count&label=stars&suffix=%20stars&color=76bad9" alt="stars">
-  <img src="https://img.shields.io/badge/dynamic/json?url=https%3A%2F%2Fapi.github.com%2Frepos%2FViperEkura%2FAstrAI&query=%24.forks_count&label=forks&suffix=%20forks&color=76bad9" alt="forks">
+  <img src="https://img.shields.io/github/v/tag/ViperEkura/AstrAI?label=Release&color=76bad9" alt="release">
+  <img src="https://img.shields.io/github/stars/ViperEkura/AstrAI?style=flat&label=Stars&color=76bad9" alt="stars">
+  <img src="https://img.shields.io/github/forks/ViperEkura/AstrAI?style=flat&label=Forks&color=76bad9" alt="forks">
 </div>
 <br>
 
@@ -28,7 +28,8 @@
 ## 📖 Table of Contents
 
 - [Features](#features)
-- [Quick Start](#quick-start)
+- [Getting Started](#getting-started)
+- [Demo](#demo)
 - [Documentation](#documentation)
 - [Contributing](#contributing)
 - [Community](#community)
@@ -49,39 +50,51 @@
 - 🤗 **HuggingFace-Style API**: AutoModel/AutoTokenizer APIs inspired by HuggingFace for easy model and tokenizer loading.
 - 🔌 **Dual API Compatibility**: Supports both OpenAI and Anthropic chat completion APIs out of the box.
 
-### Quick Start
+### Getting Started
 
-#### Installation
+End-to-end walkthrough in 5 steps:
+
+**1. Install**
 
 ```bash
 git clone https://github.com/ViperEkura/AstrAI.git
 cd AstrAI
-pip install -e .
+pip install -e .                                          # pure PyTorch (no CUDA kernels)
+# CSRC_KERNELS=true pip install -e . --no-build-isolation  # optional: fused CUDA kernels
+# pip install -e ".[dev]"                                  # dev dependencies (pytest, ruff)
 ```
 
-For development dependencies:
+**2. Download model**
 
 ```bash
-pip install -e ".[dev]"
+python scripts/demo/download.py    # downloads 1B checkpoint to params/
 ```
 
-#### Download Pre-trained Model
+**3. Preprocess data**
 
-Download pre-trained model weights (1B bilingual checkpoint) to `params/`:
+Create `pretrain.json` (preprocessing config for `seq` strategy):
+
+```json
+{
+    "version": 1,
+    "input": {"sections": [{"field": "text", "action": "train"}]},
+    "preprocessing": {"max_seq_len": 2048},
+    "output": {"storage_format": "bin"}
+}
+```
 
 ```bash
-python scripts/demo/download.py
+python scripts/tools/preprocess.py data/*.jsonl -o output/ -c pretrain.json
 ```
 
-Or download manually from [HuggingFace](https://huggingface.co/ViperEk/KHAOSZ) into `params/`.
-
-#### Train a Model
+**4. Train**
 
 ```bash
 export CUDA_VISIBLE_DEVICES=0,1,2,3
 
 nohup python scripts/tools/train.py \
     --nprocs=4 \
+    --parallel_mode=ddp \
     --train_type=seq \
     --data_root_path=/path/to/dataset \
     --param_path=/path/to/model \
@@ -90,9 +103,7 @@ nohup python scripts/tools/train.py \
     --warmup_ratio=0.05 \
     --max_lr=1e-4 \
     --max_grad_norm=1.0 \
-    --adamw_beta1=0.9 \
-    --adamw_beta2=0.95 \
-    --adamw_weight_decay=0.01 \
+    --weight_decay=0.1 \
     --window_size=2048 \
     --ckpt_interval=10000 \
     --ckpt_dir=./checkpoint \
@@ -101,15 +112,54 @@ nohup python scripts/tools/train.py \
     > out.log 2> err.log &
 ```
 
-Full reference at [Parameter Guide](assets/docs/params.md).
+**5. Serve & query**
 
-#### Generate Text
+```bash
+# Terminal 1: start server
+python scripts/tools/server.py --param_path ./params --device cuda
+
+# Terminal 2: query
+curl http://localhost:8000/v1/chat/completions \
+  -H "Content-Type: application/json" \
+  -d '{"messages":[{"role":"user","content":"Hello"}],"max_tokens":512}'
+```
+
+### Demo
+
+Check out the demos in the `scripts/demo/` folder:
+
+```bash
+# Download model weights (required before running demos)
+python scripts/demo/download.py                      # model → params/
+
+# Interactive streaming chat (multi-turn, maintains history)
+python scripts/demo/stream_chat.py
+# Type your message after >>, type !exit to quit
+
+# Batch generation (5 hardcoded prompts, non-streaming)
+python scripts/demo/generate_batch.py
+
+# Single-prompt autoregressive streaming
+python scripts/demo/generate_ar.py
+```
+
+All generation demos use `temperature=0.8`, `top_p=0.95`, `top_k=50`, `max_tokens=2048` by default and require `params/` to contain model weights (run `download.py` first).
+
+Watch a video walkthrough on [bilibili](https://www.bilibili.com/video/BV1fuLB6yEj6).
+
+---
+
+See [Documentation](#documentation) for full references beyond the examples above.
+
+#### Text Generation
+
+Batch generation from a JSONL file:
 
 ```bash
 python scripts/tools/generate.py \
-    --param_path /path/to/model \
-    --input_json_file /path/to/input.json \
-    --output_json_file /path/to/output.json
+    --param_path ./params \
+    --input_json_file input.jsonl \
+    --output_json_file output.jsonl
 ```
 
 #### Docker
@@ -123,9 +173,6 @@ docker build -t astrai:latest .
 # Run with GPU support
 docker run --gpus all -it astrai:latest
 
-# Run with specific GPUs
-docker run --gpus '"device=0,1"' -it astrai:latest
-
 # Run inference server
 docker run --gpus all -p 8000:8000 astrai:latest \
   python -m scripts.tools.server --port 8000 --device cuda
@@ -142,88 +189,42 @@ docker compose --profile cpu up -d
 
 > **Note**: `--gpus all` is required for CUDA support. Without it, `torch.cuda.is_available()` will return `False`.
 
-#### Start HTTP Server
+#### HTTP API Examples
 
-Start the inference server with OpenAI and Anthropic-compatible HTTP API:
+Additional request examples beyond the [Getting Started](#getting-started) flow:
 
 ```bash
-python -m scripts.tools.server --port 8000 --device cuda
-```
-
-Make requests:
-
-```bash
-# OpenAI-compatible
-curl -X POST http://localhost:8000/v1/chat/completions \
-  -H "Content-Type: application/json" \
-  -d '{
-    "messages": [{"role": "user", "content": "Hello"}],
-    "max_tokens": 512
-  }'
-
 # OpenAI-compatible streaming
 curl -X POST http://localhost:8000/v1/chat/completions \
   -H "Content-Type: application/json" \
-  -d '{
-    "messages": [{"role": "user", "content": "Tell a story"}],
-    "stream": true,
-    "max_tokens": 500
-  }'
+  -d '{"messages":[{"role":"user","content":"Tell a story"}],"stream":true,"max_tokens":500}'
 
 # Anthropic-compatible
 curl -X POST http://localhost:8000/v1/messages \
   -H "Content-Type: application/json" \
-  -d '{
-    "model": "astrai",
-    "system": "You are a helpful assistant.",
-    "messages": [{"role": "user", "content": "Hello"}],
-    "max_tokens": 512
-  }'
+  -d '{"model":"astrai","system":"You are a helpful assistant.","messages":[{"role":"user","content":"Hello"}],"max_tokens":512}'
 
 # Anthropic-compatible streaming with stop sequences
 curl -X POST http://localhost:8000/v1/messages \
   -H "Content-Type: application/json" \
-  -d '{
-    "model": "astrai",
-    "messages": [{"role": "user", "content": "Write a story"}],
-    "max_tokens": 500,
-    "stream": true,
-    "stop_sequences": ["The end"]
-  }'
+  -d '{"model":"astrai","messages":[{"role":"user","content":"Write a story"}],"max_tokens":500,"stream":true,"stop_sequences":["The end"]}'
 
 # Health check
 curl http://localhost:8000/health
 ```
 
-#### Demo
-
-Check out the demos in the `scripts/demo/` folder:
-
-```bash
-# Download pre‑processed data (required before running demos)
-python scripts/demo/download.py
-
-# Interactive streaming chat
-python scripts/demo/stream_chat.py
-
-# Batch generation
-python scripts/demo/generate_batch.py
-
-# Auto‑regressive generation
-python scripts/demo/generate_ar.py
-```
-
-Watch a video walkthrough on [bilibili](https://www.bilibili.com/video/BV1fuLB6yEj6).
+See [Inference Guide](assets/docs/inference.md) for SSE streaming format, error codes, and stats endpoint.
 
 ### Documentation
 
 | Document | Description |
 |----------|-------------|
-| [Parameter Guide](./assets/docs/params.md) | Training & inference parameters |
+| [CLI Reference](./assets/docs/params.md) | Parameters for all CLI tools (train, server, generate, preprocess) |
 | [Architecture](./assets/docs/architecture.md) | System architecture, class diagram & design patterns |
 | [Training](./assets/docs/training.md) | Training loop, strategies & formulas |
 | [Inference](./assets/docs/inference.md) | KVCache, continuous batching, sampling & HTTP API |
 | [Data Flow](./assets/docs/dataflow.md) | Data pipeline, storage backends & dataset architecture |
+| [Preprocessing](./assets/docs/preprocessing.md) | Declarative JSON-driven data preprocessing |
 
 ### Contributing
 

assets/docs/README-zh-CN.md

@@ -15,9 +15,9 @@
 <div align="center">
   <img src="https://img.shields.io/badge/python-3.12+-blue.svg" alt="python">
   <img src="https://img.shields.io/badge/license-GPL--3.0-blue.svg" alt="license">
-  <img src="https://img.shields.io/github/v/release/ViperEkura/AstrAI?color=76bad9" alt="release">
-  <img src="https://img.shields.io/badge/dynamic/json?url=https%3A%2F%2Fapi.github.com%2Frepos%2FViperEkura%2FAstrAI&query=%24.stargazers_count&label=stars&suffix=%20stars&color=76bad9" alt="stars">
-  <img src="https://img.shields.io/badge/dynamic/json?url=https%3A%2F%2Fapi.github.com%2Frepos%2FViperEkura%2FAstrAI&query=%24.forks_count&label=forks&suffix=%20forks&color=76bad9" alt="forks">
+  <img src="https://img.shields.io/github/v/tag/ViperEkura/AstrAI?label=Release&color=76bad9" alt="release">
+  <img src="https://img.shields.io/github/stars/ViperEkura/AstrAI?style=flat&label=Stars&color=76bad9" alt="stars">
+  <img src="https://img.shields.io/github/forks/ViperEkura/AstrAI?style=flat&label=Forks&color=76bad9" alt="forks">
 </div>
 
 <br>
@@ -34,7 +34,8 @@
 ## 📖 目录
 
 - [特性](#特性)
-- [快速开始](#快速开始)
+- [快速上手](#快速上手)
+- [演示](#演示)
 - [文档](#文档)
 - [贡献](#贡献)
 - [社区](#社区)
@@ -55,39 +56,51 @@
 - 🤗 **HuggingFace 风格 API**: 类 HuggingFace 的 AutoModel/AutoTokenizer 接口，方便加载模型和分词器。
 - 🔌 **双 API 兼容**: 同时支持 OpenAI 和 Anthropic 聊天补全 API，开箱即用。
 
-### 快速开始
+### 快速上手
 
-#### 安装
+端到端演示，只需 5 步：
+
+**1. 安装**
 
 ```bash
 git clone https://github.com/ViperEkura/AstrAI.git
 cd AstrAI
-pip install -e .
+pip install -e .                                          # 纯 PyTorch（不含 CUDA 内核）
+# CSRC_KERNELS=true pip install -e . --no-build-isolation  # 可选：融合 CUDA 内核加速
+# pip install -e ".[dev]"                                  # 可选：开发依赖（pytest, ruff）
 ```
 
-安装开发依赖：
+**2. 下载模型**
 
 ```bash
-pip install -e ".[dev]"
+python scripts/demo/download.py    # 下载 1B 检查点到 params/
 ```
 
-#### 下载预训练模型
+**3. 预处理数据**
 
-下载预训练模型权重（1B 双语检查点）到 `params/` 目录：
+创建 `pretrain.json`（`seq` 策略的预处理配置）：
+
+```json
+{
+    "version": 1,
+    "input": {"sections": [{"field": "text", "action": "train"}]},
+    "preprocessing": {"max_seq_len": 2048},
+    "output": {"storage_format": "bin"}
+}
+```
 
 ```bash
-python scripts/demo/download.py
+python scripts/tools/preprocess.py data/*.jsonl -o output/ -c pretrain.json
 ```
 
-或从 [HuggingFace](https://huggingface.co/ViperEk/KHAOSZ) 手动下载放入 `params/`。
-
-#### 训练模型
+**4. 训练**
 
 ```bash
 export CUDA_VISIBLE_DEVICES=0,1,2,3
 
 nohup python scripts/tools/train.py \
     --nprocs=4 \
+    --parallel_mode=ddp \
     --train_type=seq \
     --data_root_path=/path/to/dataset \
     --param_path=/path/to/model \
@@ -96,9 +109,7 @@ nohup python scripts/tools/train.py \
     --warmup_ratio=0.05 \
     --max_lr=1e-4 \
     --max_grad_norm=1.0 \
-    --adamw_beta1=0.9 \
-    --adamw_beta2=0.95 \
-    --adamw_weight_decay=0.01 \
+    --weight_decay=0.1 \
     --window_size=2048 \
     --ckpt_interval=10000 \
     --ckpt_dir=./checkpoint \
@@ -107,15 +118,54 @@ nohup python scripts/tools/train.py \
     > out.log 2> err.log &
 ```
 
-完整参数列表见[参数说明](./params.md)。
+**5. 启动服务并调用**
+
+```bash
+# 终端 1：启动服务
+python scripts/tools/server.py --param_path ./params --device cuda
+
+# 终端 2：发起请求
+curl http://localhost:8000/v1/chat/completions \
+  -H "Content-Type: application/json" \
+  -d '{"messages":[{"role":"user","content":"你好"}],"max_tokens":512}'
+```
+
+### 演示
+
+查看 `scripts/demo/` 文件夹中的演示：
+
+```bash
+# 下载模型权重（运行演示前必需）
+python scripts/demo/download.py                      # model → params/
+
+# 交互式流式聊天（多轮对话，保持历史记录）
+python scripts/demo/stream_chat.py
+# 在 >> 后输入消息，输入 !exit 退出
+
+# 批量生成（5 条硬编码提示词，非流式）
+python scripts/demo/generate_batch.py
+
+# 单条提示词自回归流式生成
+python scripts/demo/generate_ar.py
+```
+
+所有生成演示默认使用 `temperature=0.8`、`top_p=0.95`、`top_k=50`、`max_tokens=2048`，需要 `params/` 目录包含模型权重（请先运行 `download.py`）。
+
+观看 [bilibili](https://www.bilibili.com/video/BV1fuLB6yEj6) 上的视频演示。
+
+---
+
+更多选项请参考[文档](#文档)。
 
 #### 文本生成
 
+从 JSONL 文件批量生成：
+
 ```bash
 python scripts/tools/generate.py \
-    --param_path /path/to/model \
-    --input_json_file /path/to/input.json \
-    --output_json_file /path/to/output.json
+    --param_path ./params \
+    --input_json_file input.jsonl \
+    --output_json_file output.jsonl
 ```
 
 #### Docker
@@ -129,9 +179,6 @@ docker build -t astrai:latest .
 # 启用 GPU 运行
 docker run --gpus all -it astrai:latest
 
-# 指定特定 GPU
-docker run --gpus '"device=0,1"' -it astrai:latest
-
 # 运行推理服务
 docker run --gpus all -p 8000:8000 astrai:latest \
   python -m scripts.tools.server --port 8000 --device cuda
@@ -148,88 +195,42 @@ docker compose --profile cpu up -d
 
 > **注意**: 必须使用 `--gpus all` 才能启用 CUDA 支持，否则 `torch.cuda.is_available()` 将返回 `False`。
 
-#### 启动 HTTP 服务
+#### HTTP API 示例
 
-启动推理服务器，支持 OpenAI 和 Anthropic 兼容的 HTTP API：
+除[快速上手](#快速上手)流程外，更多请求示例：
 
 ```bash
-python -m scripts.tools.server --port 8000 --device cuda
-```
-
-发起请求：
-
-```bash
-# OpenAI 兼容
-curl -X POST http://localhost:8000/v1/chat/completions \
-  -H "Content-Type: application/json" \
-  -d '{
-    "messages": [{"role": "user", "content": "你好"}],
-    "max_tokens": 512
-  }'
-
 # OpenAI 兼容流式
 curl -X POST http://localhost:8000/v1/chat/completions \
   -H "Content-Type: application/json" \
-  -d '{
-    "messages": [{"role": "user", "content": "讲个故事"}],
-    "stream": true,
-    "max_tokens": 500
-  }'
+  -d '{"messages":[{"role":"user","content":"讲个故事"}],"stream":true,"max_tokens":500}'
 
 # Anthropic 兼容
 curl -X POST http://localhost:8000/v1/messages \
   -H "Content-Type: application/json" \
-  -d '{
-    "model": "astrai",
-    "system": "你是一个乐于助人的助手。",
-    "messages": [{"role": "user", "content": "你好"}],
-    "max_tokens": 512
-  }'
+  -d '{"model":"astrai","system":"你是一个乐于助人的助手。","messages":[{"role":"user","content":"你好"}],"max_tokens":512}'
 
 # Anthropic 兼容流式并设置停止序列
 curl -X POST http://localhost:8000/v1/messages \
   -H "Content-Type: application/json" \
-  -d '{
-    "model": "astrai",
-    "messages": [{"role": "user", "content": "写个故事"}],
-    "max_tokens": 500,
-    "stream": true,
-    "stop_sequences": ["结束"]
-  }'
+  -d '{"model":"astrai","messages":[{"role":"user","content":"写个故事"}],"max_tokens":500,"stream":true,"stop_sequences":["结束"]}'
 
 # 健康检查
 curl http://localhost:8000/health
 ```
 
-#### 演示
-
-查看 `scripts/demo/` 文件夹中的演示：
-
-```bash
-# 下载预处理数据（运行演示前必需）
-python scripts/demo/download.py
-
-# 交互式流式聊天
-python scripts/demo/stream_chat.py
-
-# 批量生成
-python scripts/demo/generate_batch.py
-
-# 自回归生成
-python scripts/demo/generate_ar.py
-```
-
-观看 [bilibili](https://www.bilibili.com/video/BV1fuLB6yEj6) 上的视频演示。
+SSE 流式格式、错误码和统计端点详见[推理文档](./inference.md)。
 
 ### 文档
 
 | 文档 | 说明 |
 |------|------|
-| [参数说明](./params.md) | 训练与推理参数配置 |
+| [CLI 参考](./params.md) | 所有 CLI 工具参数（训练、服务、生成、预处理） |
 | [架构文档](./architecture.md) | 系统架构、类图与设计模式 |
 | [训练文档](./training.md) | 训练循环、策略与公式 |
 | [推理文档](./inference.md) | KVCache、连续批处理、采样与 HTTP API |
 | [数据流程](./dataflow.md) | 数据管道、存储后端与数据集架构 |
+| [数据预处理](./preprocessing.md) | 声明式 JSON 驱动数据预处理 |
 
 ### 贡献
 

assets/docs/architecture.md

@@ -1,5 +1,12 @@
 # AstrAI Architecture
 
+## Contents
+
+- [Class Diagram](#class-diagram) — Full Mermaid class diagram across 10+ namespaces
+- [Module Overview](#module-overview) — Component inventory per module
+- [Design Patterns](#design-patterns) — 13 documented patterns with classes
+- [Core Relationships](#core-relationships) — 11 key inter-component relationships
+
 ## Class Diagram
 
 ```mermaid
@@ -8,62 +15,99 @@ classDiagram
         class BaseConfig {
             +to_dict() Dict
             +from_dict(d) Self
+            +from_file(path) Self
+            +to_file(path)
         }
 
         class BaseModelConfig {
             +Optional[str] model_type
+            +float neftune_alpha
             +from_file(config_path) Self
             +to_file(config_path)
         }
 
         class AutoRegressiveLMConfig {
-            +int vocab_size
-            +int dim
-            +int n_layers
-            +float norm_eps
-            +int dim_ffn
+            +Optional[int] vocab_size
+            +Optional[int] dim
+            +Optional[int] n_layers
+            +Optional[float] norm_eps
+            +Optional[int] dim_ffn
             +Optional[bool] tie_weight
             +Optional[dict] rope_scaling
-            +int max_len
-            +float rope_theta
+            +Optional[int] max_len
+            +Optional[float] rope_theta
             +str attn_type
-            +int n_heads
-            +int n_kv_heads
-            +bool use_qk_norm
-            +bool use_gated_attention
+            +Optional[int] n_heads
+            +Optional[int] n_kv_heads
+            +Optional[bool] use_qk_norm
+            +Optional[bool] use_gated_attention
             +Optional[int] kv_lora_rank
             +Optional[int] qk_nope_head_dim
             +Optional[int] qk_rope_head_dim
             +str ffn_type
-            +int n_routed_experts
-            +int n_shared_experts
-            +int n_activated_experts
+            +Optional[int] n_routed_experts
+            +Optional[int] n_shared_experts
+            +Optional[int] n_activated_experts
             +Optional[str] topk_method
         }
 
         class EncoderConfig {
-            +int vocab_size
-            +int dim
-            +int n_layers
-            +float norm_eps
-            +int dim_ffn
-            +int max_len
-            +float rope_theta
-            +int n_heads
-            +int n_kv_heads
-            +bool use_qk_norm
-            +bool use_gated_attention
+            +Optional[int] vocab_size
+            +Optional[int] dim
+            +Optional[int] n_layers
+            +Optional[float] norm_eps
+            +Optional[int] dim_ffn
+            +Optional[int] max_len
+            +Optional[float] rope_theta
+            +str attn_type
+            +Optional[int] n_heads
+            +Optional[int] n_kv_heads
+            +Optional[bool] use_qk_norm
+            +str ffn_type
             +Optional[dict] rope_scaling
             +Optional[str] pooling_type
             +Optional[bool] normalize_embeddings
         }
 
         class ConfigFactory {
-            +Registry _registry
+            +Dict _entries
             +register(name) decorator
             +load(raw) BaseConfig
         }
 
+        class InputConfig {
+            +Optional[List[Dict]] sections
+            +Optional[Dict[str, Dict]] sources
+        }
+
+        class ProcessingConfig {
+            +int max_seq_len
+            +int min_chars
+            +int max_chars
+            +Optional[int] max_items
+            +str packing_strategy
+            +int max_packed_len
+            +str truncation_mode
+        }
+
+        class OutputConfig {
+            +Optional[str] domain_key
+            +str storage_format
+            +int max_tokens_per_shard
+            +Dict[str, str] dtype
+            +str position_ids_mode
+        }
+
+        class PipelineConfig {
+            +int version
+            +InputConfig input
+            +dict mask
+            +str mask_default
+            +ProcessingConfig preprocessing
+            +OutputConfig output
+            +from_dict(d) Self
+        }
+
         class TrainConfig {
             +Callable[[], nn.Module] model_fn
             +str strategy
@@ -76,11 +120,10 @@ classDiagram
             +float max_grad_norm
             +list gradient_checkpointing_modules
             +int start_epoch
-            +int start_batch
+            +int start_samples
             +str ckpt_dir
             +int ckpt_interval
             +str log_dir
-            +int log_interval
             +List[str] metrics
             +Optional[LoRAConfig] lora
             +int random_seed
@@ -94,7 +137,9 @@ classDiagram
             +str start_method
             +str device_type
             +Optional[Dataset] val_dataset
+            +Optional[float] val_split
             +int val_step
+            +float neftune_alpha
             +str parallel_mode
             +dict executor_kwargs
             +dict extra_kwargs
@@ -156,13 +201,13 @@ classDiagram
         }
 
         class StoreFactory {
-            +Registry _registry
+            +Dict _entries
             +register(name) decorator
             +create(storage_type) Store
         }
 
         class DatasetFactory {
-            +Registry _registry
+            +Dict _entries
             +register(name) decorator
             +create(train_type, window_size, stride) BaseDataset
             +load(train_type, load_path, window_size, stride, storage_type) BaseDataset
@@ -173,19 +218,20 @@ classDiagram
         class Checkpoint {
             +dict state_dict
             +int epoch
-            +int iteration
+            +int consumed_samples
             +dict extra
             +dict meta
             +dict config
             +save(save_dir)
             +load(save_dir, broadcast) Checkpoint
+            +load_any(save_dir, broadcast) Optional[Checkpoint]
         }
     }
 
     namespace model {
         class AutoModel {
             +BaseModelConfig config
-            +Registry _registry
+            +Dict _entries
             +register(name) decorator
             +get_component_class(name) Type
             +from_pretrained(path, disable_random_init, strict) nn.Module
@@ -308,14 +354,44 @@ classDiagram
 
         class Embedding {
             +Parameter weight
+            +float neftune_noise_alpha
             +forward(x) Tensor
+            +set_neftune_alpha(alpha)
+        }
+    }
+
+    namespace preprocessing {
+        class BaseMaskBuilder {
+            <<abstract>>
+            +build(item, config, tokenizer) Optional[dict]
+        }
+
+        class SectionedMaskBuilder {
+            +SectionRenderer renderer
+            +build(item, config, tokenizer) Optional[dict]
+            +_build_single(item, config, tokenizer) Optional[dict]
+            +_build_multi(item, sources_spec, config, tokenizer) Optional[dict]
+        }
+
+        class Pipeline {
+            +PipelineConfig config
+            +List[str] paths
+            +str output_dir
+            +str tokenizer_path
+            +BaseMaskBuilder mask_builder
+            +PackingStrategy _packer
+            +PositionIdStrategy _position_id
+            +StoreWriter _writer
+            +transform(item) Optional[dict]
+            +run()
+            +_flush(domains, shard_idx)
         }
     }
 
     namespace tokenize {
         class AutoTokenizer {
             +vocab_size int
-            +encode(tokens, out_ids, is_pretokenized, add_special_tokens) List[int]
+            +encode(tokens, out_ids, is_pretokenized, add_special_tokens) List
             +decode(tokens, skip_special_tokens) str
             +__getattr__(name) Any (bos_id, eos_id, pad_id, stop_ids)
             +apply_chat_template(messages, system_prompt, tokenize, add_generation_prompt) Union[str, List[int]]
@@ -333,18 +409,19 @@ classDiagram
     }
 
     namespace factory {
-        class Registry {
+        class BaseFactory {
             +Dict _entries
-            +register(name, component_cls, category, priority)
-            +get(name) Type
-            +list_names() List[str]
+            +register(name) decorator
+            +create(name, *args, **kwargs) T
+            +get_component_class(name) Type
+            +list_registered() list
+            +is_registered(name) bool
         }
 
-        class BaseFactory {
-            +Registry _registry
-            +register(name, category, priority) decorator
-            +create(name, *args, **kwargs) T
-            +list_registered() list
+        class MaskBuilderFactory {
+            +Dict _entries
+            +register(name) decorator
+            +create(name, *args, **kwargs) BaseMaskBuilder
         }
     }
 
@@ -352,8 +429,8 @@ classDiagram
         class Trainer {
             +TrainConfig train_config
             +List[TrainCallback] callbacks
-            +train(checkpoint)
-            +_get_default_callbacks() List[TrainCallback]
+            +train(resume_dir)
+            -_get_default_callbacks() List[TrainCallback]
         }
 
         class TrainContext {
@@ -367,13 +444,15 @@ classDiagram
             +dict model_config
             +BaseExecutor executor
             +int epoch
-            +int iteration
+            +int consumed_samples
             +float loss
+            +float grad_norm
             +DataLoader val_dataloader
             +float val_loss
             +int world_size
             +int rank
             +dict kwargs
+            +optimizer_step() int
         }
 
         class TrainContextBuilder {
@@ -383,13 +462,17 @@ classDiagram
         }
 
         class BaseStrategy {
-            +Union[Callable, nn.Module] model
+            +Callable model
+            +Optional[BaseExecutor] executor
+            +Optional[Callable] model_fn
+            +dict extra_kwargs
             +str device
+            +__call__(batch) Tensor
             +compute_loss(batch) Tensor
         }
 
         class StrategyFactory {
-            +Registry _registry
+            +Dict _entries
             +register(name) decorator
             +create(train_type, model, device, **kwargs) BaseStrategy
         }
@@ -425,17 +508,20 @@ classDiagram
         class BaseScheduler {
             +get_lr() List[float]
             +step()
+            +state_dict() dict
+            +load_state_dict(d)
         }
 
         class SchedulerFactory {
-            +Registry _registry
+            +Dict _entries
             +register(name) decorator
-            +create(optimizer, schedule_type, **kwargs) BaseScheduler
+            +create(name, *args, **kwargs) BaseScheduler
         }
 
         class CosineScheduler {
             +int warmup_steps
             +int lr_decay_steps
+            +int total_steps
             +float min_rate
         }
 
@@ -446,6 +532,13 @@ classDiagram
             +int t_mult
         }
 
+        class WSDScheduler {
+            +int warmup_steps
+            +int stable_steps
+            +int decay_steps
+            +float min_rate
+        }
+
         class TrainCallback {
             <<protocol>>
             +on_train_begin(context)
@@ -464,7 +557,7 @@ classDiagram
         }
 
         class GradientCheckpointingCallback {
-            +tuple modules
+            +Optional[List[type]] modules
             +on_train_begin(context)
             +on_train_end(context)
         }
@@ -474,55 +567,41 @@ classDiagram
             +int interval
             +bool weight_only
             +Callable save_extra_fn
-            +_save_checkpoint(context)
+            -_save_checkpoint(context)
             +on_batch_end(context)
             +on_train_end(context)
             +on_error(context)
-            +save_extra(context)$
+            +save_extra(context) dict
         }
 
         class ProgressBarCallback {
             +int num_epoch
             +int log_interval
             +IO file
+            +tqdm progress_bar
             +on_epoch_begin(context)
-            +on_batch_end(context)
+            +on_optimizer_step(context)
             +on_epoch_end(context)
         }
 
-        class MetricLoggerCallback {
-            +str log_dir
+        class MetricCallback {
+            +Path log_dir
             +int save_interval
-            +int log_interval
             +List[str] metrics
-            +on_batch_end(context)
+            +int val_step
+            +on_optimizer_step(context)
+            +on_epoch_end(context)
             +on_train_end(context)
             +on_error(context)
-        }
-
-        class ValidationCallback {
-            +_run_validation(context)
-            +on_optimizer_step(context)
+            -_run_validation(context)
         }
 
         class CallbackFactory {
-            +Registry _registry
+            +Dict _entries
             +register(name) decorator
             +create(name, **kwargs) TrainCallback
         }
 
-        class Muon {
-            +float lr
-            +float momentum
-            +float weight_decay
-            +bool nesterov
-            +int ns_steps
-            +float adamw_lr
-            +tuple adamw_betas
-            +float adamw_eps
-            +float adamw_wd
-            +step(closure) Optional[float]
-        }
     }
 
     namespace inference {
@@ -601,20 +680,44 @@ classDiagram
         }
 
         class KVCache {
-            -PagePool _pool
-            -Storage _storage
-            -TaskTable _table
-            +int page_size
+            <<abstract>>
             +task_alloc(task_id, prompt_ids) bool
             +task_free(task_id)
             +task_extend(task_id, pos) bool
             +task_cached(task_id) int
             +task_record_hashes(task_id, prompt_ids, start_logical_page)
-            +make_table_tensor(task_ids, device) Tensor
-            +bind(page_table, total_len) KvcacheView
+            +bind_tasks(task_ids, total_len, device) CacheView
         }
 
-        class KvcacheView {
+        class PageCache {
+            +int page_size
+            -PagePool _pool
+            -Storage _storage
+            -TaskTable _table
+            +task_alloc(task_id, prompt_ids) bool
+            +task_free(task_id)
+            +task_extend(task_id, pos) bool
+            +task_cached(task_id) int
+            +task_record_hashes(task_id, prompt_ids, start_logical_page)
+            +bind_tasks(task_ids, total_len, device) PageCacheView
+        }
+
+        class ContiguousCache {
+            +int max_seq_len
+            +Tensor k, v
+            +task_alloc(task_id, prompt_ids) bool
+            +task_free(task_id)
+            +task_extend(task_id, pos) bool
+            +bind_tasks(task_ids, total_len, device) ContiguousCacheView
+        }
+
+        class CacheView {
+            <<abstract>>
+            +write(layer_id, k, v)
+            +gather(layer_id) Tuple[Tensor, Tensor]
+        }
+
+        class PageCacheView {
             -Storage _storage
             +Tensor _page_table
             +int _total_len
@@ -622,6 +725,14 @@ classDiagram
             +gather(layer_id) Tuple[Tensor, Tensor]
         }
 
+        class ContiguousCacheView {
+            -ContiguousCache _cache
+            +Tensor _batch_indices
+            +int _total_len
+            +write(layer_id, k, v)
+            +gather(layer_id) Tuple[Tensor, Tensor]
+        }
+
         class TaskTable {
             +set(task_id, page_table, cached)
             +get(task_id) List[int]
@@ -631,23 +742,22 @@ classDiagram
             +table_tensor(task_ids, device) Tensor
         }
 
-        class Task {
-            +str task_id
-            +List prompt_ids
-            +int max_tokens
-            +float temperature
-            +float top_p
-            +int top_k
-            +TaskStatus status
-            +List output_ids
-            +int input_tokens
-            +int output_tokens
-            +float arrival_time
-            +float finish_time
-            +Callable stream_callback
-            +int next_pos
-            +is_finished(stop_ids) bool
-        }
+    class Task {
+        +str task_id
+        +List prompt_ids
+        +Optional[int] max_tokens
+        +float temperature
+        +float top_p
+        +int top_k
+        +TaskStatus status
+        +List output_ids
+        +int input_tokens
+        +int output_tokens
+        +float arrival_time
+        +Optional[float] finish_time
+        +int next_pos
+        +is_finished(stop_ids) bool
+    }
 
         class TaskStatus {
             <<enumeration>>
@@ -671,6 +781,11 @@ classDiagram
             +activate(task)
             +return_to_waiting(tasks)
             +get_active_tasks() List[Task]
+            +has_work() bool
+            +wait_for_tasks(timeout)
+            +get_waiting_tasks() List[Task]
+            +clear_queues()
+            +wake()
             +get_stats() Dict
         }
 
@@ -722,7 +837,9 @@ classDiagram
 
         class ChatMessage {
             +str role
-            +str content
+            +Optional[str] content
+            +Optional[List[Dict]] tool_calls
+            +Optional[str] tool_call_id
         }
 
         class ChatCompletionRequest {
@@ -739,6 +856,8 @@ classDiagram
             +Optional[float] frequency_penalty
             +Optional[Dict[int, float]] logit_bias
             +Optional[str] user
+            +Optional[List[ToolDef]] tools
+            +Optional[Union[str, Dict]] tool_choice
         }
 
         class AnthropicMessage {
@@ -762,7 +881,7 @@ classDiagram
             <<abstract>>
             +prepare(request, engine) Tuple[str, GenContext, List[str]]
             +format_stream_start(ctx) List[str]
-            +format_chunk(token) str
+            +format_chunk(token) List[str]
             +format_stream_end(ctx, stop) List[str]
             +format_response(ctx, content, stop) Dict
         }
@@ -770,7 +889,7 @@ classDiagram
         class OpenAIResponseBuilder {
             +prepare(request, engine) Tuple
             +format_stream_start(ctx) List[str]
-            +format_chunk(token) str
+            +format_chunk(token) List[str]
             +format_stream_end(ctx, stop) List[str]
             +format_response(ctx, content, stop) Dict
         }
@@ -778,7 +897,7 @@ classDiagram
         class AnthropicResponseBuilder {
             +prepare(request, engine) Tuple
             +format_stream_start(ctx) List[str]
-            +format_chunk(token) str
+            +format_chunk(token) List[str]
             +format_stream_end(ctx, stop) List[str]
             +format_response(ctx, content, stop) Dict
         }
@@ -787,12 +906,13 @@ classDiagram
             +request
             +engine
             +builder: ResponseBuilder
-            +handle() Union[StreamingResponse, Dict]
-            -_handle_stream(agen, ctx, stops) StreamingResponse
-            -_handle_non_stream(agen, ctx, stops) Dict
+            +async handle() Union[StreamingResponse, Dict]
+            -_handle_stream(agen, ctx, stop_sequences) StreamingResponse
+            -async _handle_non_stream(agen, ctx, stop_sequences) Dict
         }
 
         class StopChecker {
+            +__init__(sequences)
             +check(text) Optional[str]
         }
 
@@ -804,9 +924,15 @@ classDiagram
             +int completion_tokens
         }
 
-        class app {
-            <<singleton>>
-            +FastAPI app
+        class StopInfo {
+            +Optional[str] matched
+            +str body
+            +str yielded
+        }
+
+        class get_app {
+            <<module>>
+            +get_app() FastAPI
         }
     }
 
@@ -829,14 +955,14 @@ classDiagram
     }
 
     namespace parallel {
-        class Functions {
+        class setup {
             <<module>>
             +spawn_parallel_fn(func, world_size, backend, master_addr, master_port, device_type, start_method, **kwargs)
-            +setup_parallel(rank, world_size, backend, master_addr, master_port, device_type)
+            +setup_parallel(rank, world_size, backend, master_addr, master_port, device_type) contextmanager
             +get_current_device() str
             +get_world_size() int
             +get_rank() int
-            +only_on_rank(rank, sync) decorator
+            +only_on_rank(rank, sync=False) decorator
         }
 
         class GradientState {
@@ -847,6 +973,7 @@ classDiagram
         class AccumOptimizer {
             +Optimizer optimizer
             +GradientState gradient_state
+            +param_groups (property)
             +step(closure)
             +zero_grad()
             +state_dict() dict
@@ -867,7 +994,7 @@ classDiagram
             +prepare(model, optimizer, dataloader, scheduler) tuple
             +accumulate(model) context manager
             +backward(loss)
-            +unwrap_model(model) nn.Module
+            +unwrap_model(model) dict
             +sync_gradients (property) bool
             +grad_accum_steps (property) int
         }
@@ -876,18 +1003,18 @@ classDiagram
         }
 
         class DDPExecutor {
-            +_prepare_model(model) nn.Module
-            +_no_sync(model) context manager
-            +unwrap_model(model) nn.Module
+            -_prepare_model(model) nn.Module
+            -_no_sync(model) context manager
+            +unwrap_model(model) dict
         }
 
         class FSDPExecutor {
-            +_prepare_model(model) nn.Module
-            +unwrap_model(model) nn.Module
+            -_prepare_model(model) nn.Module
+            +unwrap_model(model) dict
         }
 
         class ExecutorFactory {
-            +Registry _registry
+            +Dict _entries
             +register(name) decorator
             +create(parallel_mode, **kwargs) BaseExecutor
         }
@@ -899,11 +1026,25 @@ classDiagram
         }
 
         class ColumnParallelLinear {
+            +int in_features
+            +int out_features
+            +int out_features_per_rank
+            +bool gather_results
+            +Parameter weight
+            +Optional[Parameter] bias
             +forward(x) Tensor
+            +load_state_dict(state_dict)
         }
 
         class RowParallelLinear {
+            +int in_features
+            +int out_features
+            +int in_features_per_rank
+            +bool reduce_results
+            +Parameter weight
+            +Optional[Parameter] bias
             +forward(x) Tensor
+            +load_state_dict(state_dict)
         }
     }
 
@@ -916,28 +1057,32 @@ classDiagram
     BaseStrategy <|-- GRPOStrategy
     BaseScheduler <|-- CosineScheduler
     BaseScheduler <|-- SGDRScheduler
+    BaseScheduler <|-- WSDScheduler
     TrainCallback <|-- GradientClippingCallback
     TrainCallback <|-- GradientCheckpointingCallback
     TrainCallback <|-- CheckpointCallback
     TrainCallback <|-- ProgressBarCallback
-    TrainCallback <|-- MetricLoggerCallback
-    TrainCallback <|-- ValidationCallback
+    TrainCallback <|-- MetricCallback
     BaseDataset <|-- SEQDataset
     BaseDataset <|-- SFTDataset
     BaseDataset <|-- DPODataset
     BaseDataset <|-- GRPODataset
     Store <|-- H5Store
     Store <|-- MmapStore
+    Store <|-- JsonlStore
     BaseSamplingStrategy <|-- TemperatureStrategy
     BaseSamplingStrategy <|-- TopKStrategy
     BaseSamplingStrategy <|-- TopPStrategy
-    BaseSamplingStrategy <|-- SamplingPipeline
     ParallelModel <|-- RowParallelLinear
     ParallelModel <|-- ColumnParallelLinear
     AutoModel <|-- AutoRegressiveLM
     AutoModel <|-- EmbeddingEncoder
     BaseConfig <|-- BaseModelConfig
     BaseConfig <|-- TrainConfig
+    BaseConfig <|-- InputConfig
+    BaseConfig <|-- ProcessingConfig
+    BaseConfig <|-- OutputConfig
+    BaseConfig <|-- PipelineConfig
     BaseModelConfig <|-- AutoRegressiveLMConfig
     BaseModelConfig <|-- EncoderConfig
     BaseFactory <|-- AutoModel
@@ -950,16 +1095,22 @@ classDiagram
     BaseFactory <|-- StoreFactory
     BaseFactory <|-- ExecutorFactory
     BaseFactory <|-- ConfigFactory
+    BaseFactory <|-- MaskBuilderFactory
     BaseExecutor <|-- NoneExecutor
     BaseExecutor <|-- DDPExecutor
     BaseExecutor <|-- FSDPExecutor
     ResponseBuilder <|-- OpenAIResponseBuilder
     ResponseBuilder <|-- AnthropicResponseBuilder
+    BaseMaskBuilder <|-- SectionedMaskBuilder
+    KVCache <|-- PageCache
+    KVCache <|-- ContiguousCache
+    CacheView <|-- PageCacheView
+    CacheView <|-- ContiguousCacheView
 
     %% --- Composition (strong ownership, part destroyed with whole) ---
-    KVCache *-- PagePool
-    KVCache *-- Storage
-    KVCache *-- TaskTable
+    PageCache *-- PagePool
+    PageCache *-- Storage
+    PageCache *-- TaskTable
     InferenceEngine *-- InferenceScheduler
     InferenceScheduler *-- KVCache
     InferenceScheduler *-- Executor
@@ -973,7 +1124,6 @@ classDiagram
     DecoderBlock *-- RMSNorm
     ChatCompletionRequest *-- ChatMessage
     MessagesRequest *-- AnthropicMessage
-    BaseFactory *-- Registry
     BaseExecutor *-- GradientState
     AccumOptimizer o-- GradientState
     AccumScheduler o-- GradientState
@@ -988,12 +1138,17 @@ classDiagram
     TrainContext o-- BaseScheduler
     TrainContext o-- Checkpoint
     TrainContext o-- BaseExecutor
-    KvcacheView o-- Storage
+    PageCacheView o-- Storage
+    ContiguousCacheView o-- ContiguousCache
     SamplingPipeline o-- BaseSamplingStrategy
     BaseDataset o-- Store
+    Pipeline o-- PipelineConfig
+    Pipeline o-- BaseMaskBuilder
 
     %% --- Dependency (uses temporarily) ---
     TrainConfig ..> BaseStrategy : selects
+    PipelineConfig ..> MaskBuilderFactory : selects
+    MaskBuilderFactory ..> BaseMaskBuilder : creates
     StrategyFactory ..> BaseStrategy : creates
     SchedulerFactory ..> BaseScheduler : creates
     DatasetFactory ..> BaseDataset : creates
@@ -1006,6 +1161,7 @@ classDiagram
     DecoderBlock ..> FFNFactory : uses
     StoreFactory ..> H5Store : creates
     StoreFactory ..> MmapStore : creates
+    StoreFactory ..> JsonlStore : creates
     ConfigFactory ..> AutoRegressiveLMConfig : creates
     ConfigFactory ..> EncoderConfig : creates
     ExecutorFactory ..> NoneExecutor : creates
@@ -1019,7 +1175,8 @@ classDiagram
     TrainContextBuilder ..> ResumableDistributedSampler : creates
     Checkpoint ..> Checkpoint : serializes
     CheckpointCallback ..> Checkpoint : creates
-    KVCache ..> KvcacheView : binds
+    PageCache ..> PageCacheView : binds
+    ContiguousCache ..> ContiguousCacheView : binds
     InferenceEngine ..> GenerationRequest : uses
     InferenceEngine ..> GenerateResult : creates
     OpenAIResponseBuilder ..> ChatCompletionRequest : receives
@@ -1046,15 +1203,16 @@ classDiagram
 
 | Module | Components | Description |
 |--------|------------|-------------|
-| **astrai.config** | BaseConfig, BaseModelConfig, AutoRegressiveLMConfig, EncoderConfig, ConfigFactory, TrainConfig | Configuration management (to_dict/from_dict, to_file/from_file) |
-| **astrai.dataset** | BaseDataset–GRPODataset, Store–MmapStore, StoreFactory, ResumableDistributedSampler, DatasetFactory | Dataset loading and management |
+| **astrai.config** | BaseConfig, BaseModelConfig, AutoRegressiveLMConfig, EncoderConfig, ConfigFactory, TrainConfig, PipelineConfig, InputConfig, ProcessingConfig, OutputConfig | Configuration management (to_dict/from_dict, to_file/from_file) |
+| **astrai.preprocessing** | BaseMaskBuilder, MaskBuilderFactory, SectionedMaskBuilder, Pipeline, filter_by_length, PackingStrategy, PackingStrategyFactory, PositionIdStrategy, PositionIdStrategyFactory, StoreWriter, StoreWriterFactory | Declarative JSON-driven data preprocessing |
+| **astrai.dataset** | BaseDataset–GRPODataset, Store–JsonlStore/MmapStore/H5Store, StoreFactory, ResumableDistributedSampler, DatasetFactory | Dataset loading and management |
 | **astrai.serialization** | Checkpoint | Model serialization |
 | **astrai.model** | AutoModel, AutoRegressiveLM, EmbeddingEncoder, DecoderBlock, GQA, MLA, MLP, DeepSeekMoE, AttnFactory, FFNFactory, RMSNorm, Linear, RotaryEmbedding, Embedding | Neural network model |
 | **astrai.tokenize** | AutoTokenizer, ChatTemplate | Tokenizer and chat template |
-| **astrai.trainer** | Trainer, TrainContext, TrainContextBuilder, BaseStrategy–GRPOStrategy, StrategyFactory, BaseScheduler–SGDRScheduler, SchedulerFactory, TrainCallback(Protocol)–ValidationCallback, CallbackFactory, Muon | Training workflow |
-| **astrai.inference** | InferenceEngine, InferenceScheduler, Executor, KVCache–KvcacheView, Allocator–Storage, Task, TaskManager, TaskStatus, GenerationRequest, GenerateResult, BaseSamplingStrategy–SamplingPipeline, ProtocolHandler, ResponseBuilder, OpenAIResponseBuilder, AnthropicResponseBuilder, StopChecker, GenContext, ChatMessage–MessagesRequest, app | Inference service |
+| **astrai.trainer** | Trainer, TrainContext, TrainContextBuilder, BaseStrategy–GRPOStrategy, StrategyFactory, BaseScheduler–WSDScheduler, SchedulerFactory, TrainCallback(Protocol)–MetricCallback, CallbackFactory | Training workflow |
+| **astrai.inference** | InferenceEngine, InferenceScheduler, Executor, KVCache–ContiguousCache/PageCache, CacheView–ContiguousCacheView/PageCacheView, Allocator–Storage, Task, TaskManager, TaskStatus, GenerationRequest, GenerateResult, BaseSamplingStrategy–SamplingPipeline, ProtocolHandler, ResponseBuilder, OpenAIResponseBuilder, AnthropicResponseBuilder, StopChecker, GenContext, ChatMessage–MessagesRequest, app | Inference service |
 | **astrai.parallel** | spawn_parallel_fn, setup_parallel, get_rank/get_world_size/get_current_device, only_on_rank, BaseExecutor, ExecutorFactory, NoneExecutor, DDPExecutor, FSDPExecutor, GradientState, AccumOptimizer, AccumScheduler, ParallelModel, RowParallelLinear, ColumnParallelLinear | Distributed parallel & gradient accumulation |
-| **astrai.factory** | Registry, BaseFactory[T] | Component registration |
+| **astrai.factory** | BaseFactory | Component registration |
 | **astrai.protocols** | OptimizerProtocol, SchedulerProtocol | Structural subtyping for optimizer/scheduler wrappers |
 
 ## Design Patterns
@@ -1062,7 +1220,7 @@ classDiagram
 | Pattern | Classes | Purpose |
 |---------|---------|---------|
 | **Factory** | `AttnFactory`, `FFNFactory`, `StrategyFactory`, `DatasetFactory`, `SchedulerFactory`, `CallbackFactory`, `StoreFactory`, `ConfigFactory`, `ExecutorFactory` | Decorator-based component creation |
-| **Registry** | `BaseFactory`, `Registry` | Component registration with category/priority |
+| **Registry** | `BaseFactory` | Component registration |
 | **Strategy** | `SEQStrategy`, `SFTStrategy`, `DPOStrategy`, `GRPOStrategy` | Training strategy switching |
 | **Strategy (Sampling)** | `TemperatureStrategy`, `TopKStrategy`, `TopPStrategy`, `SamplingPipeline` | Composable logit transformations |
 | **Strategy (API)** | `ResponseBuilder`, `OpenAIResponseBuilder`, `AnthropicResponseBuilder` | HTTP API handler with format hooks |
@@ -1070,23 +1228,23 @@ classDiagram
 | **Observer** | `TrainCallback`, callback implementations | Training process monitoring |
 | **Context** | `TrainContext` | Unified training state bag |
 | **Object Pool** | `Allocator`, `PagePool` | Page-based KV cache with LRU eviction |
-| **Executor** | `BaseExecutor`, `NoneExecutor`, `DDPExecutor` | Gradient accumulation & model distribution |
-| **Storage** | `Store`, `H5Store`, `MmapStore` | Format-agnostic data access with multi-segment support |
+| **Executor** | `BaseExecutor`, `NoneExecutor`, `DDPExecutor`, `FSDPExecutor` | Gradient accumulation & model distribution |
+| **Storage** | `Store`, `H5Store`, `MmapStore`, `JsonlStore` | Format-agnostic data access with multi-segment support |
 | **Producer-Consumer** | `InferenceScheduler`, `Task`, queues | Continuous batching |
 | **AutoModel Registry** | `AutoModel`, `AutoRegressiveLM`, `EmbeddingEncoder` | Model-type dynamic loading |
 
 ## Core Relationships
 
-1. **Config → Training**: `TrainConfig` holds model, dataset, optimizer_fn, scheduler_fn, `parallel_mode`, `executor_kwargs`
+1. **Config → Training**: `TrainConfig` holds `model_fn`, `dataset`, `optimizer_fn`, `scheduler_fn`, `parallel_mode`, `executor_kwargs`
 2. **Training Flow**: `Trainer` → `TrainContextBuilder` → `TrainContext`, uses `BaseStrategy` for loss, `BaseExecutor` for gradient accumulation + model distribution
 3. **Strategy Selection**: `StrategyFactory` creates strategy by `train_type`
 4. **Executor Selection**: `ExecutorFactory.create(cfg.parallel_mode, grad_accum_steps=cfg.grad_accum_steps, **cfg.executor_kwargs)` → `NoneExecutor` / `DDPExecutor` / `FSDPExecutor`
 5. **Inference Flow**: `InferenceEngine` → `InferenceScheduler` → `AutoRegressiveLM`, backed by `KVCache` + `SamplingPipeline`
 6. **Distributed**: `spawn_parallel_fn` + `setup_parallel` for multi-process DDP
-7. **Dataset Loading**: `DatasetFactory` creates datasets, `Store` (H5Store/MmapStore) loads data with explicit `_length` and multi-segment `_data`
+7. **Dataset Loading**: `DatasetFactory` creates datasets, `Store` (H5Store/MmapStore/JsonlStore) loads data with explicit `_length` and multi-segment `_data`
 8. **Checkpoint**: `Checkpoint` saves/loads safetensors + metadata (rank-0 only), extra state saved as `{key}.pt`
-9. **Scheduler**: `SchedulerFactory` creates `CosineScheduler`/`SGDRScheduler`
+9. **Scheduler**: `SchedulerFactory` creates `CosineScheduler`/`SGDRScheduler`/`WSDScheduler`
 10. **AutoModel**: `from_pretrained()` loads `config.json` + `model.safetensors`, `_disable_random_init` replaces `nn.init.*` with no-ops
 11. **Protocols**: `OptimizerProtocol` / `SchedulerProtocol` — structural subtyping for `AccumOptimizer` / `AccumScheduler` wrappers
 
-> Document Update Time: 2026-05-28
+> Document Update Time: 2026-07-09

assets/docs/dataflow.md

@@ -1,46 +1,101 @@
 # Data Flow
 
-This document describes the data pipeline: from raw text to model input tensors.
+This document describes the data pipeline: from raw text to model input tensors. For creating preprocessing configs, see [Preprocessing Guide](preprocessing.md).
+
+## Contents
+
+- [Overview](#overview)
+- [Data Preparation](#data-preparation) — tokenization, format detection, backends
+- [Data Keys by Training Type](#data-keys-by-training-type)
+- [Dataset Architecture](#dataset-architecture)
+- [Sampler](#sampler)
+- [DataLoader](#dataloader)
 
 ## Overview
 
 ```
-Raw Text → AutoTokenizer → Token IDs → .h5/.bin → Dataset → Sampler → DataLoader → Training/Inference
+JSONL Lines → Pipeline (mask builder) → Tokenized Tensors
+                                              ↓
+                                      .h5 or .bin storage
+                                              ↓
+                                      Store.load()
+                                              ↓
+                                      Store.fetch(begin, end, keys)
+                                              ↓
+                                      BaseDataset.__getitem__(idx)
+                                              ↓
+                                      Sampler → DataLoader → Training / Inference
 ```
 
 ## Data Preparation
 
 Raw text is tokenized via `AutoTokenizer.encode()` and saved as HDF5 (`.h5`) or binary (`.bin` + `meta.json`) files with keyed tensor groups.
 
+### Tokenization
+
+The `Pipeline` reads JSONL lines, applies the mask builder (see [Preprocessing](preprocessing.md)), and produces flat token sequences:
+
+```python
+# Per JSONL line: messages → chat template → token IDs + loss mask
+tokens = tokenizer.encode(rendered_text)        # List[int]
+loss_mask = [0, 0, 0, 1, 1, 1, 1, 1, 1]        # 0=masked, 1=train
+# Stored as flat tensors, packed with other lines by packing strategy
+```
+
+The output `meta.json` records the storage format, key names, dtype, total token count, and tensor shapes for each shard.
+
+### Format Detection
+
+`detect_format(load_path)` inspects the path:
+
+- If `load_path` is a file: checks suffix — `.h5`/`.hdf5` → `"h5"`, `.jsonl` → `"jsonl"`, unknown suffix raises `ValueError`
+- If `load_path` is a directory: recursively globs for `*.h5`/`*.hdf5` files → `"h5"`, `*.bin` + `**/meta.json` → `"bin"`, or `*.jsonl` + `dataset_config.json` → `"jsonl"`
+
+### Store Backends
+
 Storage format is auto-detected by `detect_format()`; backends are dispatched via registry:
 
 ```
-StoreFactory.create("h5")  → H5Store
-StoreFactory.create("bin") → MmapStore
+StoreFactory.create("h5")    → H5Store
+StoreFactory.create("bin")   → MmapStore
+StoreFactory.create("jsonl") → JsonlStore
 ```
 
-H5 backend supports shared memory via `.share_memory_()`. Bin (mmap) uses OS page-cache sharing natively.
+**H5Store**: Reads HDF5 files. Tensors are loaded into host memory and normalized into segmented storage.
+
+**MmapStore**: Memory-maps `.bin` files. OS page cache sharing is native — no explicit `share_memory_()` needed. Uses `torch.from_numpy(np.memmap(...))`.
+
+**JsonlStore**: On-the-fly tokenization of raw JSONL files at load time. Requires a `dataset_config.json` alongside the `.jsonl` files following the same `PipelineConfig` schema with an additional `tokenizer_path` field.
+
+All backends normalise tensors into `Store._data[Dict[str, List[Tensor]]]` + `Store._cum[Dict[str, List[int]]]` (cumulative lengths for bisect-based indexing).
 
 ## Data Keys by Training Type
 
 | Type | Storage Keys |
 |------|-------------|
 | `seq` | `sequence` (→ input_ids, target_ids via offset-by-1) |
-| `sft` | `sequence`, `loss_mask` |
+| `sft` | `sequence`, `loss_mask`, `position_ids` |
 | `dpo` | `chosen`, `rejected`, `chosen_mask`, `rejected_mask` |
 | `grpo` | `prompts`, `responses`, `masks`, `rewards` |
 
 ## Dataset Architecture
 
 ```
-DatasetFactory.load(train_type, load_path, window_size, stride, storage_type)
-  → StoreFactory.create(detect_format(path))
-    → Store._data[Dict[str, List[Tensor]]] + _cum[Dict[str, List[int]]]
-      → BaseDataset.__getitem__(idx)
-        → sliding window [begin, end) via get_index(idx)
+DatasetFactory.load(train_type, load_path, window_size, stride=None, storage_type=None)
+  → BaseDataset.load(load_path, storage_type=None)
+    → detect_format(load_path)
+    → StoreFactory.create(storage_type)
+    → Store.load(load_path)
+      → _normalize(raw)  # base Store, shared by both backends
+        → Store._data[Dict[str, List[Tensor]]] + _cum[Dict[str, List[int]]]
+          → BaseDataset.__getitem__(idx)
+            → get_index(idx) → [begin, end)
+            → Store.fetch(begin, end, keys) → Tensor / Dict[str, Tensor]
 ```
 
-`window_size` = max input length, `stride` = step between consecutive samples (defaults to `window_size`).
+`window_size` = max input length, `stride` = step between consecutive samples (defaults to `window_size`, optional). `storage_type` defaults to `None` (auto-detect via `detect_format`).
+
+`Store.fetch(begin, end, keys)` accepts a single key (`str`) returning a `Tensor`, or a list of keys returning `Dict[str, Tensor]`. Internally uses `bisect` across multi-segment tensors. Raises `RuntimeError("Store not loaded")` if called before `load()`.
 
 ## Sampler
 
@@ -54,4 +109,4 @@ DatasetFactory.load(train_type, load_path, window_size, stride, storage_type)
 
 Standard PyTorch `DataLoader` with configurable `batch_size`, `num_workers`, `pin_memory`, `prefetch_factor`. Sampler produces indices; dataloader fetches tensor batches via `__getitem__`.
 
-> Document Update Time: 2026-05-28
+> Document Update Time: 2026-07-09

assets/docs/inference.md

@@ -1,5 +1,16 @@
 # Inference
 
+## Contents
+
+- [KV Cache](#kv-cache)
+- [KVCache System](#kvcache-system)
+- [Continuous Batching](#continuous-batching)
+- [Sampling](#sampling-strategy-pattern)
+- [Protocol Handlers](#protocol-handlers-strategy-pattern)
+- [Engine & GenerateResult](#engine--generateresult)
+- [HTTP API](#http-api) — endpoints, SSE, errors, stats
+- [Engine API](#engine-api)
+
 ## KV Cache
 
 At decode time, only the last query token matters. All previous K/V are cached to avoid recomputation:
@@ -12,29 +23,40 @@ RoPE is applied **before** KV cache write, not after — otherwise position enco
 
 ## KVCache System
 
-Six classes working together:
+Seven classes working together, with two concrete cache implementations:
+
+### ContiguousCache (default)
 
 ```
-KVCache (facade)
-  ├── PagePool         orchestrates page allocation + prefix matching
-  │     ├── Allocator   bitmask-based page allocator + ref-count + LRU eviction (inside PagePool)
-  │     └── PrefixCache hash-based prefix matching (page_hash via polynomial hash) (inside PagePool)
-  ├── TaskTable        maps task_id → page_table + cached token count
-  ├── Storage          k_cache / v_cache tensors (n_layers × n_pages × page_size × n_kv_heads × head_dim)
-  └── KvcacheView      bundles Storage + page_table + total_len for attention layers (returned by bind())
+ContiguousCache (simple contiguous per-slot cache)
+  ├── ContiguousCacheView  bundles k/v tensors + slot indices for attention layers
 ```
 
-`KVCache.bind(page_table, total_len)` returns a `KvcacheView` used by attention layers via `write()` / `gather()`.
+Created by default when no cache is passed to `InferenceScheduler`. Each task occupies a fixed slot of `[max_seq_len, n_kv_heads, head_dim]`. Simple and efficient for small-to-medium batch sizes.
+
+### PageCache (paged with prefix sharing)
+
+```
+PageCache (paged KV cache with prefix sharing, alternative)
+  ├── PagePool               orchestrates page allocation + prefix matching
+  │     ├── Allocator         bitmask-based page allocator + ref-count + LRU
+  │     └── PrefixCache       hash-based prefix matching (page_hash via polynomial hash)
+  ├── TaskTable              maps task_id → page_table + cached token count
+  ├── Storage                k_cache / v_cache tensors (n_layers × n_pages × page_size × n_kv_heads × head_dim)
+  └── PageCacheView          bundles Storage + page_table + total_len for attention layers
+```
+
+`isinstance(cache, KVCache)` checks dispatch to the correct view. Both implement the abstract `KVCache` interface used by `Executor` and `InferenceScheduler`.
 
 ## Continuous Batching
 
 `InferenceScheduler` runs a daemon thread with a 4-phase loop:
 
 ```
-1. Cleanup → Remove finished tasks, free KV pages
-2. Refill  → Pop from waiting_queue, task_alloc pages, activate
+1. Cleanup → Remove finished tasks, free KV cache slots/pages
+2. Refill  → Pop from waiting_queue, task_alloc resources, activate
 3. Prefill → Group by (prompt_len, start_pos), run full forward
-4. Decode  → Pick largest same-position group, single-token forward
+4. Decode  → Run single-token forward for each same-position group
 ```
 
 ## Sampling (Strategy Pattern)
@@ -43,7 +65,8 @@ KVCache (facade)
 BaseSamplingStrategy (ABC)
   ├── TemperatureStrategy
   ├── TopKStrategy
-  └── TopPStrategy
+  ├── TopPStrategy
+  └── SamplingPipeline
 ```
 
 `SamplingPipeline` composes them: Temperature → Top-K → Top-P → softmax → multinomial.  
@@ -73,7 +96,9 @@ Adding a protocol = one builder file, no handler subclassing needed.
 InferenceEngine
   ├── generate(prompt, stream, ...) → str | List[str] | Generator
   ├── generate_with_request(req)    → same
-  └── generate_async(prompt, ...)   → AsyncGenerator
+  ├── generate_async(prompt, ...)   → AsyncGenerator
+  ├── get_stats()                   → Dict
+  └── shutdown()
 ```
 
 `GenerateResult` uses `Condition` for non-streaming (`wait_completion()`) and `Event` for streaming (`wait()`). Stream callback is `cb(token)`.
@@ -124,12 +149,90 @@ Supports `stop_sequences` and streaming via `event: content_block_delta`.
 | Param | Type | Default | Description |
 |-------|------|---------|-------------|
 | `messages` | List[dict] | required | Chat messages (role, content) |
-| `temperature` | float | 1.0 | Sampling temperature (>= 0.0) |
-| `top_p` | float | 1.0 | Nucleus threshold |
 | `top_k` | int | 50 | Top-k count |
+| `top_p` | float | 1.0 | Nucleus threshold |
+| `temperature` | float | 1.0 | Sampling temperature (> 0.0) |
 | `max_tokens` | Optional[int] | None | Max generation length |
 | `stream` | bool | False | Stream output |
 
+### SSE Streaming Format
+
+**OpenAI** (`/v1/chat/completions`, `stream=true`):
+
+```
+data: {"id":"chatcmpl-...","object":"chat.completion.chunk","created":...,"model":"astrai",
+       "choices":[{"index":0,"delta":{"role":"assistant"},"finish_reason":null}]}
+
+data: {"id":"chatcmpl-...","object":"chat.completion.chunk","created":0,"model":"astrai",
+       "choices":[{"index":0,"delta":{"content":"Hello"},"finish_reason":null}]}
+
+data: {"id":"chatcmpl-...","object":"chat.completion.chunk","created":...,"model":"astrai",
+       "choices":[{"index":0,"delta":{},"finish_reason":"stop"}]}
+
+data: {"prompt_tokens":5,"completion_tokens":1,"total_tokens":6}
+
+data: [DONE]
+```
+
+**Anthropic** (`/v1/messages`, `stream=true`):
+
+```
+event: message_start
+data: {"type":"message_start","message":{"id":"msg_...","model":"astrai","role":"assistant",
+       "content":[],"usage":{"input_tokens":0}}}
+
+event: content_block_start
+data: {"type":"content_block_start","index":0,"content_block":{"type":"text","text":""}}
+
+event: content_block_delta
+data: {"type":"content_block_delta","index":0,"delta":{"type":"text_delta","text":"Hello"}}
+
+event: content_block_stop
+data: {"type":"content_block_stop","index":0}
+
+event: message_delta
+data: {"type":"message_delta","delta":{"stop_reason":"end_turn","stop_sequence":null},"usage":{...}}
+
+event: message_stop
+data: {"type":"message_stop"}
+```
+
+### Error Responses
+
+The server returns standard HTTP status codes. Pydantic validation errors (e.g. missing required fields)
+are handled automatically by FastAPI with 422 status. The only application-level error is engine initialization:
+
+| Status | Meaning |
+|--------|---------|
+| 200 | Success |
+| 422 | Unprocessable entity (Pydantic validation) |
+| 503 | Service unavailable (model not loaded, engine not ready) |
+
+Error response body (503):
+
+```json
+{
+    "detail": "Engine not initialized"
+}
+```
+
+### Stats Endpoint
+
+```
+GET /stats
+```
+
+Response:
+
+```json
+{
+    "total_tasks": 128,
+    "total_tokens": 10240,
+    "active_tasks": 3,
+    "waiting_queue": 2
+}
+```
+
 ## Engine API
 
 ```python
@@ -142,7 +245,8 @@ engine.generate("Hello", stream=True)           # -> Generator[str]
 engine.generate(["A", "B"], stream=True)        # -> Generator[Tuple[int, str]]
 
 # Async
-await engine.generate_async("Hello", ...)       # -> AsyncGenerator[str]
+async for token in engine.generate_async("Hello", ...):    # -> AsyncGenerator[str]
+    print(token)
 ```
 
-> Document Update Time: 2026-05-28
+> Document Update Time: 2026-07-09

assets/docs/params.md

@@ -1,4 +1,11 @@
-# Parameter Documentation
+# CLI Parameter Reference
+
+## Contents
+
+- [Training Parameters](#training-parameters)
+- [Inference Server](#inference-server-serverpy)
+- [Generate](#generate-generatepy)
+- [Preprocess](#preprocess-preprocesspy)
 
 ## Training Parameters
 
@@ -21,13 +28,17 @@
 | `--max_lr` | Maximum learning rate (cosine decay after warmup) | 3e-4 |
 | `--max_grad_norm` | Maximum gradient norm for clipping | 1.0 |
 
-### Optimizer (AdamW)
+### Optimizer (MuonMix)
+
+Combined optimizer: matrix parameters via **Muon**, non-matrix via **AdamW** (`fused=True`).
 
 | Parameter | Description | Default |
 |-----------|-------------|---------|
-| `--adamw_beta1` | AdamW beta1 | 0.9 |
-| `--adamw_beta2` | AdamW beta2 | 0.95 |
-| `--adamw_weight_decay` | AdamW weight decay | 0.01 |
+| `--weight_decay` | Weight decay (applied to Muon matrix params; non-matrix use 0) | 0.1 |
+| `--muon_momentum` | Muon momentum factor | 0.95 |
+| `--muon_nesterov` | Enable Nesterov momentum for Muon | True |
+| `--muon_ns_steps` | Newton-Schulz iteration steps for Muon | 5 |
+| `--muon_adjust_lr` | Muon LR adjustment strategy (`original`, `match_rms_adamw`) | `match_rms_adamw` |
 
 ### Data Loading
 
@@ -46,7 +57,27 @@
 | `--ckpt_interval` | Iterations between checkpoints | 5000 |
 | `--ckpt_dir` | Checkpoint save directory | checkpoint |
 | `--start_epoch` | Resume from epoch (0 = from scratch) | 0 |
-| `--start_batch` | Resume from batch iteration | 0 |
+| `--start_samples` | Resume from sample count per rank | 0 |
+
+### Validation
+
+| Parameter | Description | Default |
+|-----------|-------------|---------|
+| `--val_split` | Ratio to split from training dataset for validation (e.g. 0.05) | None |
+| `--val_step` | Number of optimizer steps between validation runs | 1000 |
+
+### Logging
+
+| Parameter | Description | Default |
+|-----------|-------------|---------|
+| `--log_dir` | Directory for metric logs | checkpoint/logs |
+| `--metrics` | Metrics to log (e.g. --metrics loss lr val_loss) | ["loss", "lr", "grad_norm"] |
+
+### Gradient Checkpointing
+
+| Parameter | Description | Default |
+|-----------|-------------|---------|
+| `--gradient_checkpointing` | Enable activation checkpointing for DecoderBlock modules | False |
 
 ### Distributed Training
 
@@ -56,17 +87,32 @@
 | `--parallel_mode` | Parallel strategy (`none`, `ddp`, or `fsdp`) | none |
 | `--device_type` | Device type | cuda |
 | `--start_method` | Multiprocessing start method (`spawn`, `fork`, `forkserver`) | spawn |
+| `--backend` | Distributed training backend | nccl |
+| `--master_addr` | Master node address | localhost |
+| `--master_port` | Master node port | 29500 |
 
 ### Strategy-specific
 
 | Parameter | Description | Default | Used by |
 |-----------|-------------|---------|---------|
 | `--dpo_beta` | DPO beta value | 0.1 | `dpo` |
-| `--label_smoothing` | Label smoothing for cross-entropy loss | 0.05 | `seq`, `sft` |
+| `--label_smoothing` | Label smoothing for cross-entropy loss | 0.0 | `seq`, `sft` |
 | `--group_size` | GRPO group size | 4 | `grpo` |
 | `--grpo_clip_eps` | GRPO clipping epsilon | 0.2 | `grpo` |
 | `--grpo_kl_coef` | GRPO KL penalty coefficient | 0.01 | `grpo` |
 | `--grpo_sync_interval` | GRPO ref_model sync interval (steps) | 200 | `grpo` |
+| `--neftune_alpha` | NEFTune noise alpha (0=disabled, typical: 5.0) | 0.0 | `sft` |
+
+### Scheduler
+
+| Parameter | Description | Default |
+|-----------|-------------|---------|
+| `--schedule_type` | LR scheduler type (`cosine`, `sgdr`, `wsd`) | cosine |
+| `--min_rate` | Minimum LR as fraction of base LR | None (scheduler default: 0.01) |
+| `--cycle_length` | SGDR first cycle length in steps | None (total_steps - warmup_steps) |
+| `--t_mult` | SGDR cycle length multiplier per restart | 2 |
+| `--stable_steps` | WSD stable plateau steps | None (required for wsd) |
+| `--decay_steps` | WSD decay steps | None (total_steps - warmup_steps - stable_steps) |
 
 ### Usage Example
 
@@ -75,6 +121,7 @@ export CUDA_VISIBLE_DEVICES=0,1,2,3
 
 nohup python scripts/tools/train.py \
     --nprocs=4 \
+    --parallel_mode=ddp \
     --train_type=seq \
     --data_root_path=/path/to/dataset \
     --param_path=/path/to/model \
@@ -83,9 +130,7 @@ nohup python scripts/tools/train.py \
     --warmup_ratio=0.05 \
     --max_lr=1e-4 \
     --max_grad_norm=1.0 \
-    --adamw_beta1=0.9 \
-    --adamw_beta2=0.95 \
-    --adamw_weight_decay=0.01 \
+    --weight_decay=0.1 \
     --window_size=2048 \
     --ckpt_interval=10000 \
     --ckpt_dir=./checkpoint \
@@ -96,4 +141,64 @@ nohup python scripts/tools/train.py \
 
 ---
 
-> Document Update Time: 2026-05-24
+## Inference Server (`server.py`)
+
+| Parameter | Type | Default | Description |
+|-----------|------|---------|-------------|
+| `--host` | str | `0.0.0.0` | Host address |
+| `--port` | int | `8000` | Port number |
+| `--param_path` | path | `project_root/params` | Path to model parameters |
+| `--device` | str | `cuda` | Device to load model on |
+| `--dtype` | str | `bfloat16` | Model weights dtype (`bfloat16`, `float16`, `float32`) |
+| `--max_batch_size` | int | `16` | Maximum batch size for continuous batching |
+| `--reload` | flag | `False` | Enable auto-reload for development |
+
+Usage:
+```bash
+python scripts/tools/server.py --param_path ./params --device cuda --dtype bfloat16
+```
+
+See [Inference Guide](inference.md) for HTTP API documentation.
+
+## Generate (`generate.py`)
+
+| Parameter | Type | Default | Description |
+|-----------|------|---------|-------------|
+| `--param_path` | str | required | Path to the model directory |
+| `--input_json_file` | str | required | Path to the input JSONL file |
+| `--output_json_file` | str | required | Path to the output JSONL file |
+| `--question_key` | str | `question` | Key for the question in input JSON |
+| `--response_key` | str | `response` | Key for the response in output JSON |
+| `--temperature` | float | `0.60` | Sampling temperature |
+| `--top_k` | int | `30` | Top-k filtering |
+| `--top_p` | float | `0.95` | Nucleus sampling threshold |
+| `--batch_size` | int | `1` | Batch size for generation |
+| `--max_tokens` | int | model config `max_len` | Maximum tokens to generate |
+
+Usage:
+```bash
+python scripts/tools/generate.py \
+    --param_path ./params \
+    --input_json_file input.jsonl \
+    --output_json_file output.jsonl
+```
+
+## Preprocess (`preprocess.py`)
+
+| Parameter | Type | Default | Description |
+|-----------|------|---------|-------------|
+| `input_files` | path(s) | required | Input JSONL file(s), supports glob (`data/*.jsonl`) |
+| `--output_dir`, `-o` | path | required | Output directory for processed data |
+| `--config`, `-c` | path | required | Preprocessing pipeline config (JSON) |
+| `--tokenizer_path` | str | `params` | Path to tokenizer directory |
+
+Usage:
+```bash
+python scripts/tools/preprocess.py data/*.jsonl -o output/ -c sft.json
+```
+
+See [Preprocessing Guide](preprocessing.md) for config file format and examples.
+
+---
+
+> Document Update Time: 2026-07-09

assets/docs/preprocessing.md

@@ -0,0 +1,364 @@
+# Preprocessing Pipeline
+
+Declarative JSON-driven data preprocessing. One `SectionedMaskBuilder` handles all formats via `input.sections` (single-output) or `input.sources` (multi-output).
+
+## Contents
+
+- [Philosophy](#philosophy)
+- [Config Structure](#config-structure)
+- [Quick Start](#quick-start) — SFT Chat, SFT Instruction, Pretrain, DPO, GRPO examples
+- [Configuration Reference](#configuration-reference) — all fields
+- [Mask Algorithm](#mask-algorithm)
+- [Output Layout](#output-layout)
+- [CLI](#cli)
+- [Python API](#python-api)
+
+## Philosophy
+
+| Component | Responsibility |
+|-----------|---------------|
+| `tokenizer_config.json` (`chat_template`) | Formatting -- how roles become tokens |
+| `pipeline.json` (`mask`) | Masking -- which roles participate in training |
+
+A single config file captures the entire pipeline, reusable and version-controllable.
+
+## Config Structure
+
+```json
+{
+  "version":       1,
+  "input":         {},   // sections (single) or sources (multi)
+  "mask":          {},   // role -> "train" | "mask"
+  "mask_default":  "mask",
+  "preprocessing": {},
+  "output":        {}
+}
+```
+
+### Section Fields
+
+| Field | Type | Default | Description |
+|-------|------|---------|-------------|
+| `field` | str | -- | JSONL key to read |
+| `action` | str | -- | `"train"` / `"mask"` / `"$role"` |
+| `template` | bool | `false` | Apply `chat_template` per message |
+| `add_special_tokens` | bool | `true` for first non-template section | Add special tokens during encode |
+
+### Source Fields (multi-output mode)
+
+| Field | Type | Default | Description |
+|-------|------|---------|-------------|
+| `sections` | list[dict] | -- | Same as single-output section list |
+| `list_field` | bool | `false` | JSONL field holds a list; tokenise each element |
+| `mask_key` | str | `"{key}_mask"` | Explicit output key for loss mask |
+
+---
+
+## Quick Start
+
+### SFT Chat
+
+Input JSONL:
+
+```json
+{"messages": [{"role": "system", "content": "You are helpful."}, {"role": "user", "content": "Hi"}, {"role": "assistant", "content": "Hello!"}]}
+```
+
+Config:
+
+```json
+{
+  "input": {
+    "sections": [
+      {"field": "messages", "action": "$role", "template": true}
+    ]
+  },
+  "mask": {
+    "system": "mask",
+    "user": "mask",
+    "assistant": "train"
+  },
+  "mask_default": "mask",
+  "preprocessing": {
+    "max_seq_len": 2048
+  },
+  "output": {
+    "storage_format": "bin",
+    "dtype": {"loss_mask": "bool"}
+  }
+}
+```
+
+Output keys: `sequence` (int32), `loss_mask` (bool)
+
+### SFT Instruction
+
+Input JSONL:
+
+```json
+{"prompt": "Translate to French: Hello", "response": "Bonjour"}
+```
+
+Config:
+
+```json
+{
+  "input": {
+    "sections": [
+      {"field": "prompt",   "action": "mask", "add_special_tokens": true},
+      {"field": "response", "action": "train"}
+    ]
+  },
+  "mask_default": "mask",
+  "preprocessing": {
+    "max_seq_len": 2048
+  }
+}
+```
+
+Output keys: `sequence`, `loss_mask`
+
+### Pretrain
+
+Input JSONL:
+
+```json
+{"text": "Artificial Intelligence is a field of computer science..."}
+```
+
+Config:
+
+```json
+{
+  "input": {
+    "sections": [
+      {"field": "text", "action": "train"}
+    ]
+  },
+  "preprocessing": {
+    "max_seq_len": 8192,
+    "min_chars": 100
+  }
+}
+```
+
+Output keys: `sequence` (no `loss_mask` — all tokens trained)
+
+### DPO
+
+Input JSONL:
+
+```json
+{"chosen": [{"role": "user", "content": "What is 2+2?"}, {"role": "assistant", "content": "4"}], "rejected": [{"role": "user", "content": "What is 2+2?"}, {"role": "assistant", "content": "5"}]}
+```
+
+Config:
+
+```json
+{
+  "input": {
+    "sources": {
+      "chosen": {
+        "sections": [
+          {"field": "chosen", "action": "$role", "template": true}
+        ]
+      },
+      "rejected": {
+        "sections": [
+          {"field": "rejected", "action": "$role", "template": true}
+        ]
+      }
+    }
+  },
+  "mask": {
+    "user": "mask",
+    "assistant": "train"
+  },
+  "mask_default": "mask"
+}
+```
+
+Output keys: `chosen`, `chosen_mask`, `rejected`, `rejected_mask`
+
+### GRPO
+
+Input JSONL:
+
+```json
+{"prompt": [{"role": "user", "content": "What is 2+2?"}], "responses": ["4", "Five", "Four"], "rewards": [1.0, 0.3, 0.8]}
+```
+
+Config:
+
+```json
+{
+  "input": {
+    "sources": {
+      "prompts": {
+        "sections": [
+          {"field": "prompt", "action": "mask", "template": true}
+        ]
+      },
+      "responses": {
+        "sections": [
+          {"field": "responses", "action": "train"}
+        ],
+        "list_field": true,
+        "mask_key": "masks"
+      },
+      "rewards": {
+        "sections": [
+          {"field": "rewards", "action": "value"}
+        ]
+      }
+    }
+  },
+  "mask": {
+    "user": "mask",
+    "assistant": "train"
+  },
+  "mask_default": "mask"
+}
+```
+
+Output keys: `prompts`, `prompts_mask`, `responses`, `masks`, `rewards` (float32)
+
+- `action: "value"` — extract raw values from JSONL without tokenisation
+- `list_field: true` — tokenise each list element independently, then concatenate
+- `mask_key: "masks"` — rename the auto-generated mask key (default: `responses_mask`)
+- `prompts_mask` is auto-generated (all masked) and unused by GRPOStrategy
+
+---
+
+## Configuration Reference
+
+### `input`
+
+| Field | Type | Default | Description |
+|-------|------|---------|-------------|
+| `sections` | list[dict] or null | `null` | Section specs for single-output mode |
+| `sources` | dict[str, dict] or null | `null` | Source specs for multi-output mode (DPO/GRPO) |
+
+When `sources` is set, `sections` is ignored.
+
+### `mask`
+
+| Field | Type | Default | Description |
+|-------|------|---------|-------------|
+| `mask` | dict | `{}` | `{role: "train" \| "mask"}` |
+| `mask_default` | str | `"mask"` | Default action for unlisted roles |
+
+### `preprocessing`
+
+| Field | Type | Default | Description |
+|-------|------|---------|-------------|
+| `max_seq_len` | int | `2048` | Truncate sequences to this length |
+| `min_chars` | int | `50` | Skip text-mode items shorter than this |
+| `max_chars` | int | `2000000` | Skip text-mode items longer than this |
+| `max_items` | int or null | `null` | Stop after N documents |
+| `packing_strategy` | str | `"simple"` | Packing strategy: `"simple"`, `"bfd"`, `"bfd_split"` |
+| `max_packed_len` | int | `8192` | Maximum length of a packed bin |
+| `truncation_mode` | str | `"keep_start"` | How to truncate sequences: `"keep_start"` or `"keep_end"` |
+
+### `output`
+
+| Field | Type | Default | Description |
+|-------|------|---------|-------------|
+| `domain_key` | str or null | `null` | JSONL key for domain grouping |
+| `storage_format` | str | `"bin"` | `"bin"` (mmap) or `"h5"` |
+| `max_tokens_per_shard` | int | `100000000` | Flush threshold in cumulative tokens |
+| `dtype` | dict[str, str] | `{}` | Per-key tensor dtype override (e.g. `{"loss_mask": "bool"}`) |
+| `position_ids_mode` | str | `"doc_reset"` | How to compute position_ids: `"none"`, `"doc_reset"`, `"continuous"` |
+
+---
+
+## Mask Algorithm
+
+### Template mode (`template: true`)
+
+1. Prepend BOS token (masked)
+2. For each message in the field's array:
+   1. Render through `chat_template` for that single message
+   2. Encode rendered text
+   3. Apply mask rule for the message's role
+
+### Non-template mode
+
+Encode the field value as text. Mask value is 1 (train) or 0 (mask) per the section's `action`.
+
+### Text config detection
+
+When no section uses `template` and all sections have `action: "train"`, the builder omits `loss_mask` from the output — all tokens are trained.
+
+---
+
+## Output Layout
+
+### Single-Shard (`bin`)
+
+```
+output/
+  __default__/
+    shard_0000/
+      meta.json
+      sequence.bin
+      loss_mask.bin
+  wiki/
+    shard_0000/
+      meta.json
+      sequence.bin
+      loss_mask.bin
+```
+
+### Multi-Shard (`bin`)
+
+When `max_tokens_per_shard` is exceeded:
+
+```
+output/
+  __default__/
+    shard_0000/
+      meta.json
+      sequence.bin
+      loss_mask.bin
+    shard_0001/
+      meta.json
+      sequence.bin
+      loss_mask.bin
+```
+
+For `bin` format, `MmapStore` discovers all shards under the domain directory via `rglob("meta.json")`. For `h5` format, `H5Store` discovers `.h5`/`.hdf5` files via recursive glob.
+
+---
+
+## CLI
+
+```bash
+# SFT
+python scripts/tools/preprocess.py data/sft/*.jsonl -o output/sft/ -c configs/sft_chat.json
+
+# DPO
+python scripts/tools/preprocess.py data/dpo/*.jsonl -o output/dpo/ -c configs/dpo.json --tokenizer_path params
+
+# GRPO
+python scripts/tools/preprocess.py data/grpo/*.jsonl -o output/grpo/ -c configs/grpo.json
+```
+
+---
+
+## Python API
+
+```python
+from astrai.preprocessing.pipeline import Pipeline
+from astrai.config.preprocess_config import PipelineConfig
+
+config = PipelineConfig.from_file("sft.json")
+Pipeline(
+    config,
+    ["data_part1.jsonl", "data_part2.jsonl"],
+    output_dir="output/",
+    tokenizer_path="params",
+).run()
+```
+
+> Document Update Time: 2026-07-09

assets/docs/training.md

@@ -1,37 +1,17 @@
 # Training
 
-## Model Architecture
+## Contents
 
-The model uses a decoder-only Transformer with **GQA** (Grouped Query Attention) and optional **MLA** (Multi-head Latent Attention). 1.0 billion parameters, Chinese–English bilingual.
-
-```mermaid
-flowchart TB
-    subgraph Layers["Transformer Layers"]
-        direction TB
-        A[Input Embedding] --> B[Transformer Block\nLayer 1]
-        B --> C[Transformer Block\nLayer ...]
-        C --> D[Transformer Block\nLayer ...]
-        D --> E[RMSNorm]
-        E --> F[Linear]
-        F --> G[SoftMax]
-    end
-
-    subgraph TransformerBlock["Transformer Block"]
-        direction TB
-        H[x] --> I[RMSNorm]
-        I --> J[Linear → Q/K/V]
-        J --> K[Q]; J --> L[K]; J --> M[V]
-        K --> N[RoPE]; L --> O[RoPE]
-        N --> P["Q @ K^T / sqrt(d)"]; O --> P
-        P --> Q[Masked SoftMax]; Q --> R[S @ V]; M --> R
-        R --> S[Linear]; S --> T[+]; H --> T
-        T --> U[RMSNorm]
-        U --> V["Linear (gate)"]; U --> W["Linear (up)"]
-        V --> X[SiLU]; X --> Y[×]; W --> Y
-        Y --> Z["Linear (down)"]; Z --> AA[+]; T --> AA
-        AA --> BB[x']
-    end
-```
+- [Autoregression](#autoregression)
+- [Causal Mask](#causal-mask)
+- [Rotary Position Embedding (RoPE)](#rotary-position-embedding-rope)
+- [Training Loop](#training-loop)
+- [Strategies](#strategies) — SEQ, SFT, DPO, GRPO
+- [LR Schedulers](#lr-schedulers)
+- [Gradient Checkpointing](#gradient-checkpointing)
+- [Checkpoint](#checkpoint)
+- [TrainContextBuilder](#traincontextbuilder-builder-pattern)
+- [Training CLI](#training-cli)
 
 ### Autoregression
 
@@ -69,14 +49,18 @@ Two-level loop: **epoch** → **batch**. Optimizer step fires every `grad_accum_
 
 ```
 on_train_begin
+  model.train()
   on_epoch_begin
     for batch in dataloader:
       on_batch_begin
       with executor.accumulate(model):
-        loss = strategy(batch)
+        loss = strategy.compute_loss(batch)
+        context.loss = loss.item()
         stand_loss = loss / executor.grad_accum_steps
         executor.backward(stand_loss)
-        iteration += 1
+        context.consumed_samples += (
+            context.config.batch_per_device * context.world_size
+        )
         on_batch_end
 
         if executor.sync_gradients:
@@ -94,11 +78,15 @@ on_train_end
 | Hook | Fires | Default callback |
 |------|-------|-----------------|
 | `on_train_begin` | Before training starts | `GradientCheckpointingCallback` |
-| `on_optimizer_step` | Every accumulation window | `GradientClippingCallback`, `ValidationCallback` |
-| `on_batch_end` | Every batch | `CheckpointCallback`, `MetricLoggerCallback`, `ProgressBarCallback` |
-| `on_train_end` | Training ends | `CheckpointCallback`, `MetricLoggerCallback` (final save) |
+| `on_epoch_begin` | Start of each epoch | `ProgressBarCallback` |
+| `on_batch_begin` | Every batch | — |
+| `on_optimizer_step` | Every accumulation window | `GradientClippingCallback`, `MetricCallback`, `ProgressBarCallback` |
+| `on_batch_end` | Every batch | `CheckpointCallback` |
+| `on_epoch_end` | End of each epoch | `MetricCallback`, `ProgressBarCallback` |
+| `on_error` | On exception during training | `CheckpointCallback`, `MetricCallback` |
+| `on_train_end` | Training ends (always via finally) | `CheckpointCallback`, `MetricCallback`, `GradientCheckpointingCallback` |
 
-Default callbacks (in order): `gradient_checkpointing` (activation checkpointing, optional), `checkpoint` (safetensors, rank-0), `metric_logger` (JSONL, rank-0), `progress_bar` (tqdm), `gradient_clipping`, `validation` (periodic validation on val_dataset).
+Default callbacks (in order): `gradient_checkpointing` (activation checkpointing, optional), `checkpoint` (safetensors, rank-0), `metric` (JSONL + validation, rank-0), `progress_bar` (tqdm), `gradient_clipping`.
 
 ## Strategies
 
@@ -110,7 +98,7 @@ $$
 L_{\text{PT}} = -\sum_{t=1}^{T} \log P(x_t \mid x_{\lt t}; \theta)
 $$
 
-Keys: `input_ids`, `target_ids`
+Keys: `input_ids`, `target_ids`. Optional: `label_smoothing`.
 
 ### SFT (Supervised Fine-Tuning)
 
@@ -120,7 +108,7 @@ $$
 L_{\text{SFT}} = -\sum_{t=P+1}^{P+L} \log P(s_t \mid s_{\lt t}; \theta)
 $$
 
-Keys: `input_ids`, `target_ids`, `loss_mask`
+Keys: `input_ids`, `target_ids`, `loss_mask`, `position_ids`. Optional: `label_smoothing`.
 
 ### DPO (Direct Preference Optimization)
 
@@ -130,7 +118,7 @@ $$
 L_{\text{DPO}} = -\mathbb{E}\left[\log\sigma\left(\beta\log\frac{\pi_\theta(y_w\mid x)}{\pi_{\text{ref}}(y_w\mid x)} - \beta\log\frac{\pi_\theta(y_l\mid x)}{\pi_{\text{ref}}(y_l\mid x)}\right)\right]
 $$
 
-Parameters: `beta=0.1`. Keys: `chosen`, `rejected`, `chosen_mask`, `rejected_mask`.
+Parameters: `beta=0.1`, `reduction="mean"`. Keys: `chosen`, `rejected`, `chosen_mask`, `rejected_mask`.
 
 ### GRPO (Group Relative Policy Optimization)
 
@@ -144,7 +132,7 @@ $$
 L_{\text{GRPO}} = -\mathbb{E}\left[\min\left(\frac{\pi_\theta}{\pi_{\text{ref}}}A,\; \text{clip}\left(\frac{\pi_\theta}{\pi_{\text{ref}}}, 1-\epsilon, 1+\epsilon\right)A\right)\right] + \lambda \cdot \mathbb{E}\left[(\log\pi_\theta - \log\pi_{\text{ref}})^2\right]
 $$
 
-Parameters: `group_size=4`, `clip_eps=0.2`, `kl_coef=0.01`, `sync_interval=200`.
+Parameters: `group_size=4`, `clip_eps=0.2`, `kl_coef=0.01`, `sync_interval=200`, `reduction="mean"`.
 
 Keys: `prompts`, `responses`, `masks`, `rewards`.
 
@@ -154,8 +142,9 @@ Keys: `prompts`, `responses`, `masks`, `rewards`.
 |------|-------|-------------|
 | Cosine | `CosineScheduler` | Linear warmup → cosine decay to `min_rate` |
 | SGDR | `SGDRScheduler` | Cosine annealing with warm restarts (`t_mult=2`) |
+| WSD | `WSDScheduler` | Warmup-Stable-Decay with sqrt cooldown |
 
-Created by `SchedulerFactory.create(optimizer, schedule_type, **kwargs)`.
+Created by `SchedulerFactory.create(schedule_type, optimizer, **kwargs)`. Valid types: `"cosine"`, `"sgdr"`, `"wsd"`. Omit to use no scheduler.
 
 ## Gradient Checkpointing
 
@@ -171,9 +160,9 @@ Callback wraps each `DecoderBlock.forward` with `torch.utils.checkpoint.checkpoi
 ## Checkpoint
 
 ```
-Checkpoint(state_dict, epoch, iteration, extra, meta, config)
-  ├── save(save_dir)    rank-0 only: meta.json (epoch/iteration/timestamp) + config.json (model config) + state_dict.safetensors + optional {key}.pt (optimizer.pt, scheduler.pt)
-  └── load(save_dir)    broadcasts metadata from rank-0
+Checkpoint(state_dict, epoch, consumed_samples, extra, meta, config)
+  ├── save(save_dir)    rank-0 only: meta.json (epoch/consumed_samples/timestamp) + config.json (model config) + model.safetensors + optional {key}.pt (optimizer.pt, scheduler.pt)
+  └── load(save_dir, broadcast=False)    loads from local disk; set broadcast=True to broadcast metadata from rank-0
 ```
 
 Optimizer/scheduler state persisted by default via `Checkpoint.extra`.  
@@ -194,7 +183,7 @@ context = (
 - Creates executor via `ExecutorFactory.create(cfg.parallel_mode, grad_accum_steps=cfg.grad_accum_steps, **cfg.executor_kwargs)`
 - Calls `executor.prepare(model, optimizer, dataloader, scheduler)` for model distribution (e.g. DDP) + gradient accumulation wrappers
 - Creates `ResumableDistributedSampler` for shuffle+resume
-- Builds strategy via `StrategyFactory.create(train_type, ...)`
+- Builds strategy via `StrategyFactory.create(train_type, model, device, **kwargs)`
 
 ## Training CLI
 
@@ -203,6 +192,7 @@ export CUDA_VISIBLE_DEVICES=0,1,2,3
 
 nohup python scripts/tools/train.py \
     --nprocs=4 \
+    --parallel_mode=ddp \
     --train_type=seq \
     --data_root_path=/path/to/dataset \
     --param_path=/path/to/model \
@@ -211,9 +201,7 @@ nohup python scripts/tools/train.py \
     --warmup_ratio=0.05 \
     --max_lr=1e-4 \
     --max_grad_norm=1.0 \
-    --adamw_beta1=0.9 \
-    --adamw_beta2=0.95 \
-    --adamw_weight_decay=0.01 \
+    --weight_decay=0.1 \
     --window_size=2048 \
     --ckpt_interval=10000 \
     --ckpt_dir=./checkpoint \
@@ -224,4 +212,4 @@ nohup python scripts/tools/train.py \
 
 Full parameter reference at [params.md](params.md).
 
-> Document Update Time: 2026-05-28
+> Document Update Time: 2026-07-09

astrai/__init__.py

@@ -1,34 +1,98 @@
-__version__ = "1.3.7"
+__version__ = "1.3.8"
 __author__ = "ViperEkura"
 
 from astrai.config import (
     AutoRegressiveLMConfig,
+    BaseModelConfig,
+    ConfigFactory,
     EncoderConfig,
+    PipelineConfig,
     TrainConfig,
 )
-from astrai.dataset import DatasetFactory
+from astrai.dataset import (
+    BaseDataset,
+    DatasetFactory,
+    ResumableDistributedSampler,
+    Store,
+    StoreFactory,
+)
 from astrai.factory import BaseFactory
 from astrai.inference import (
     GenerationRequest,
     InferenceEngine,
+    ProtocolHandler,
+    SamplingPipeline,
+    get_app,
+    run_server,
+    sample,
+)
+from astrai.model import (
+    AutoModel,
+    AutoRegressiveLM,
+    EmbeddingEncoder,
+    LoRAConfig,
+    inject_lora,
+)
+from astrai.parallel import (
+    ExecutorFactory,
+    get_rank,
+    get_world_size,
+    only_on_rank,
+    spawn_parallel_fn,
+)
+from astrai.preprocessing import Pipeline, filter_by_length
+from astrai.serialization import Checkpoint
+from astrai.tokenize import AutoTokenizer, ChatTemplate
+from astrai.trainer import (
+    BaseScheduler,
+    BaseStrategy,
+    CallbackFactory,
+    SchedulerFactory,
+    StrategyFactory,
+    TrainCallback,
+    Trainer,
 )
-from astrai.model import AutoModel, AutoRegressiveLM
-from astrai.tokenize import AutoTokenizer
-from astrai.trainer import CallbackFactory, SchedulerFactory, StrategyFactory, Trainer
 
 __all__ = [
     "AutoRegressiveLM",
     "AutoRegressiveLMConfig",
-    "EncoderConfig",
-    "TrainConfig",
-    "DatasetFactory",
+    "AutoModel",
     "AutoTokenizer",
+    "BaseDataset",
+    "BaseFactory",
+    "BaseModelConfig",
+    "BaseScheduler",
+    "BaseStrategy",
+    "CallbackFactory",
+    "ChatTemplate",
+    "Checkpoint",
+    "ConfigFactory",
+    "DatasetFactory",
+    "EmbeddingEncoder",
+    "EncoderConfig",
+    "ExecutorFactory",
     "GenerationRequest",
     "InferenceEngine",
-    "Trainer",
-    "CallbackFactory",
-    "StrategyFactory",
+    "LoRAConfig",
+    "Pipeline",
+    "PipelineConfig",
+    "ProtocolHandler",
+    "ResumableDistributedSampler",
+    "SamplingPipeline",
     "SchedulerFactory",
-    "BaseFactory",
-    "AutoModel",
+    "Store",
+    "StoreFactory",
+    "StrategyFactory",
+    "TrainCallback",
+    "TrainConfig",
+    "Trainer",
+    "filter_by_length",
+    "get_app",
+    "get_rank",
+    "get_world_size",
+    "inject_lora",
+    "only_on_rank",
+    "run_server",
+    "sample",
+    "spawn_parallel_fn",
 ]

astrai/config/__init__.py

@@ -4,13 +4,22 @@ from astrai.config.model_config import (
     ConfigFactory,
     EncoderConfig,
 )
+from astrai.config.preprocess_config import (
+    InputConfig,
+    OutputConfig,
+    PipelineConfig,
+    ProcessingConfig,
+)
 from astrai.config.train_config import TrainConfig
 
 __all__ = [
-    # Model configuration
     "BaseModelConfig",
     "AutoRegressiveLMConfig",
     "EncoderConfig",
     "ConfigFactory",
     "TrainConfig",
+    "InputConfig",
+    "OutputConfig",
+    "PipelineConfig",
+    "ProcessingConfig",
 ]

astrai/config/base.py

@@ -1,6 +1,7 @@
 import json
 from dataclasses import MISSING, dataclass, fields
-from typing import Any, Dict, Optional, Self, get_type_hints
+from pathlib import Path
+from typing import Any, Dict, Optional, Self, Union, get_type_hints
 
 
 @dataclass
@@ -83,4 +84,15 @@ class BaseConfig:
             return value
         if isinstance(value, target_type):
             return value
+        if isinstance(value, dict) and issubclass(target_type, BaseConfig):
+            return target_type.from_dict(value)
         raise TypeError
+
+    @classmethod
+    def from_file(cls, path: Union[str, Path]) -> Self:
+        with open(path, "r", encoding="utf-8") as f:
+            return cls.from_dict(json.load(f))
+
+    def to_file(self, path: Union[str, Path]):
+        with open(path, "w", encoding="utf-8") as f:
+            json.dump(self.to_dict(), f, indent=2, ensure_ascii=False)

astrai/config/model_config.py

@@ -1,6 +1,5 @@
-import json
 from dataclasses import dataclass
-from typing import Any, Dict, Optional, Self
+from typing import Any, Dict, Optional
 
 from astrai.config.base import BaseConfig
 from astrai.factory import BaseFactory
@@ -21,18 +20,7 @@ class BaseModelConfig(BaseConfig):
     """Base config with ``model_type`` dispatch and file I/O."""
 
     model_type: Optional[str] = None
-
-    @classmethod
-    def from_file(cls, config_path: str) -> Self:
-        with open(config_path, "r") as f:
-            raw: Dict[str, Any] = json.load(f)
-        return cls.from_dict(raw)
-
-    def to_file(self, config_path: str):
-        d = self.to_dict()
-        config_dict = {k: v for k, v in d.items() if v is not None}
-        with open(config_path, "w") as f:
-            json.dump(config_dict, f, indent=4)
+    neftune_alpha: float = 0.0
 
 
 @dataclass
@@ -83,10 +71,12 @@ class EncoderConfig(BaseModelConfig):
     rope_theta: Optional[float] = None
     rope_scaling: Optional[dict] = None
 
+    attn_type: str = "gqa"
     n_heads: Optional[int] = None
     n_kv_heads: Optional[int] = None
     use_qk_norm: Optional[bool] = None
     use_gated_attention: Optional[bool] = None
 
+    ffn_type: str = "mlp"
     pooling_type: Optional[str] = None
     normalize_embeddings: Optional[bool] = None

astrai/config/preprocess_config.py

@@ -0,0 +1,109 @@
+"""Pipeline configuration for JSONL preprocessing.
+
+Supports single-sequence (SFT/pretrain) and multi-output (DPO/GRPO)
+modes, both driven declaratively through ``input.sections`` or
+``input.sources``.
+"""
+
+from dataclasses import dataclass, field
+from typing import Dict, List, Optional
+
+from astrai.config.base import BaseConfig
+
+
+@dataclass
+class InputConfig(BaseConfig):
+    """Declarative input mapping.
+
+    Single-output mode (backward-compatible)::
+
+        {"input": {"sections": [{"field": "messages", ...}]}}
+
+    Multi-output mode (DPO / GRPO)::
+
+        {"input": {"sources": {
+            "chosen": {"sections": [{"field": "chosen", ...}]},
+            "rejected": {"sections": [{"field": "rejected", ...}]},
+        }}}
+    """
+
+    sections: Optional[List[Dict]] = None
+    sources: Optional[Dict[str, Dict]] = None
+
+
+@dataclass
+class ProcessingConfig(BaseConfig):
+    """Processing configuration.
+
+    Parameters
+    ----------
+    max_seq_len : int
+        Maximum sequence length (default: 2048).
+    min_chars : int
+        Minimum number of characters to keep (default: 50).
+    max_chars : int
+        Maximum number of characters to keep (default: 2_000_000).
+    max_items : Optional[int]
+        Maximum number of items to process (default: None, unlimited).
+    packing_strategy : str
+        How to pack sequences into a contiguous stream.
+
+        - ``"simple"``: sequential concatenation (default, backward compatible).
+        - ``"bfd"``: best-fit decreasing bin packing, minimises wasted tokens.
+        - ``"bfd_split"``: BFD with over-length sequences split into chunks.
+    max_packed_len : int
+        Maximum length of a packed bin. Sequences longer than this are
+        truncated or split depending on ``packing_strategy`` (default: 8192).
+    truncation_mode : str
+        How to truncate sequences longer than ``max_packed_len``.
+
+        - ``"keep_start"``: keep the first ``max_packed_len`` tokens (default).
+        - ``"keep_end"``: keep the last ``max_packed_len`` tokens.
+    """
+
+    max_seq_len: int = 2048
+    min_chars: int = 50
+    max_chars: int = 2_000_000
+    max_items: Optional[int] = None
+    packing_strategy: str = "simple"
+    max_packed_len: int = 8192
+    truncation_mode: str = "keep_start"
+
+
+@dataclass
+class OutputConfig(BaseConfig):
+    """Output configuration.
+
+    Parameters
+    ----------
+    domain_key : Optional[str]
+        Domain key for the output store (default: None).
+    storage_format : str
+        Storage format, one of ``"bin"``, ``"jsonl"`` (default: ``"bin"``).
+    max_tokens_per_shard : int
+        Maximum tokens per shard before splitting (default: 100_000_000).
+    dtype : Dict[str, str]
+        Per-key dtype overrides, e.g. ``{"input_ids": "int32"}`` (default: {}).
+    position_ids_mode : Optional[str]
+        How to compute position_ids in packed sequences.
+
+         - ``"none"``: do not generate (default).
+        - ``"doc_reset"``: reset to 0 at each document boundary.
+        - ``"continuous"``: sequential 0, 1, 2, ... (pretrain, single doc).
+    """
+
+    domain_key: Optional[str] = None
+    storage_format: str = "bin"
+    max_tokens_per_shard: int = 100_000_000
+    dtype: Dict[str, str] = field(default_factory=dict)
+    position_ids_mode: str = "doc_reset"
+
+
+@dataclass
+class PipelineConfig(BaseConfig):
+    version: int = 1
+    input: InputConfig = field(default_factory=InputConfig)
+    mask: Dict[str, str] = field(default_factory=dict)
+    mask_default: str = "mask"
+    preprocessing: ProcessingConfig = field(default_factory=ProcessingConfig)
+    output: OutputConfig = field(default_factory=OutputConfig)

astrai/config/train_config.py

@@ -1,5 +1,5 @@
 from dataclasses import dataclass, field, fields
-from typing import Callable, List, Optional
+from typing import Any, Callable, Dict, List, Optional
 
 import torch.nn as nn
 from torch.optim import Optimizer
@@ -40,21 +40,25 @@ class TrainConfig(BaseConfig):
     max_grad_norm: float = field(
         default=1.0, metadata={"help": "Maximum gradient norm."}
     )
-    gradient_checkpointing_modules: list = field(
+    gradient_checkpointing_modules: List[str] = field(
         default_factory=list,
         metadata={"help": "Module types to enable activation checkpointing for."},
     )
 
     # checkpoint setting
     start_epoch: int = field(default=0, metadata={"help": "Start epoch for training."})
-    start_batch: int = field(
-        default=0, metadata={"help": "Start batch iteration for training."}
+    start_samples: int = field(
+        default=0,
+        metadata={
+            "help": "Start samples count (per rank). Superseded by checkpoint consumed_samples."
+        },
     )
     ckpt_dir: str = field(
         default="./checkpoint", metadata={"help": "Checkpoint directory."}
     )
     ckpt_interval: int = field(
-        default=5000, metadata={"help": "Number of iterations between checkpoints."}
+        default=5000,
+        metadata={"help": "Number of optimizer steps between checkpoints."},
     )
 
     # lora setting
@@ -67,12 +71,8 @@ class TrainConfig(BaseConfig):
     log_dir: str = field(
         default="./checkpoint/logs", metadata={"help": "Directory for metric logs."}
     )
-    log_interval: int = field(
-        default=100,
-        metadata={"help": "Number of batch iterations between metric logs."},
-    )
     metrics: List[str] = field(
-        default_factory=lambda: ["loss", "lr"],
+        default_factory=lambda: ["loss", "lr", "grad_norm"],
         metadata={"help": "Metrics to record during training."},
     )
 
@@ -118,16 +118,26 @@ class TrainConfig(BaseConfig):
     val_dataset: Optional[Dataset] = field(
         default=None, metadata={"help": "Dataset for validation."}
     )
+    val_split: Optional[float] = field(
+        default=None,
+        metadata={
+            "help": "Ratio to split from training dataset for validation (e.g. 0.05). Ignored if val_dataset is set."
+        },
+    )
     val_step: int = field(
         default=1000,
         metadata={"help": "Number of optimizer steps between validation runs."},
     )
+    neftune_alpha: float = field(
+        default=0.0,
+        metadata={"help": "NEFTune noise alpha (0=disabled, typical: 5.0)."},
+    )
 
-    executor_kwargs: dict = field(
+    executor_kwargs: Dict[str, Any] = field(
         default_factory=dict,
         metadata={"help": "Extra kwargs passed to ExecutorFactory.create()."},
     )
-    extra_kwargs: dict = field(
+    extra_kwargs: Dict[str, Any] = field(
         default_factory=dict, metadata={"help": "Other arguments."}
     )
 

astrai/dataset/__init__.py

@@ -5,10 +5,13 @@ from astrai.dataset.dataset import (
 from astrai.dataset.sampler import ResumableDistributedSampler
 from astrai.dataset.storage import (
     H5Store,
+    JsonlStore,
     MmapStore,
     Store,
     StoreFactory,
     detect_format,
+)
+from astrai.serialization import (
     load_bin,
     load_h5,
     save_bin,
@@ -22,6 +25,7 @@ __all__ = [
     "StoreFactory",
     "H5Store",
     "MmapStore",
+    "JsonlStore",
     "detect_format",
     "save_h5",
     "load_h5",

astrai/dataset/dataset.py

@@ -48,24 +48,26 @@ class BaseDataset(Dataset, ABC):
                 f"Missing: {missing}"
             )
 
-    def load(self, load_path: str, storage_type: Optional[str] = None):
+    def load(self, load_path: str, storage_type: Optional[str] = None, **kwargs):
         """Load dataset from the given path.
 
         Auto-detects the storage format if not specified.
 
         Args:
             load_path: Path to the data directory or file
-            storage_type: Force a specific storage type ("h5", "bin"),
+            storage_type: Force a specific storage type ("h5", "bin", "jsonl"),
                           or None for auto-detection
+            **kwargs: Extra arguments forwarded to the store constructor and
+                      to ``store.load()``.
 
         Raises:
             KeyError: If the loaded storage is missing required keys.
         """
         if storage_type is None:
             storage_type = detect_format(load_path)
-        self.storage = StoreFactory.create(storage_type)
+        self.storage = StoreFactory.create(storage_type, **kwargs)
         self._load_path = load_path
-        self.storage.load(load_path)
+        self.storage.load(load_path, **kwargs)
         self._validate_keys()
 
     @property
@@ -136,26 +138,6 @@ class DatasetFactory(BaseFactory["BaseDataset"]):
         dataset = DatasetFactory.create("custom", window_size, stride)
     """
 
-    @classmethod
-    def _validate_component(cls, dataset_cls: type):
-        """Validate that the dataset class inherits from BaseDataset."""
-        if not issubclass(dataset_cls, BaseDataset):
-            raise TypeError(f"{dataset_cls.__name__} must inherit from BaseDataset")
-
-    @classmethod
-    def create(cls, train_type: str, window_size: int, stride: int) -> "BaseDataset":
-        """Create a dataset instance.
-
-        Args:
-            train_type: Type of training ("seq", "sft", "dpo", "grpo")
-            window_size: Window size for data sampling
-            stride: Stride between consecutive samples
-
-        Returns:
-            Dataset instance
-        """
-        return super().create(train_type, window_size, stride)
-
     @classmethod
     def load(
         cls,
@@ -164,6 +146,7 @@ class DatasetFactory(BaseFactory["BaseDataset"]):
         window_size: int,
         stride: Optional[int] = None,
         storage_type: Optional[str] = None,
+        **kwargs,
     ) -> "BaseDataset":
         """Create and load a dataset in one step.
 
@@ -172,7 +155,8 @@ class DatasetFactory(BaseFactory["BaseDataset"]):
             load_path: Path to the data file
             window_size: Window size for data sampling
             stride: Stride between consecutive samples (default: same as window_size)
-            storage_type: Storage type ("h5", "bin") or None for auto-detection
+            storage_type: Storage type ("h5", "bin", "jsonl") or None for auto-detection
+            **kwargs: Extra arguments forwarded to ``dataset.load()``.
 
         Returns:
             Loaded dataset instance
@@ -181,23 +165,15 @@ class DatasetFactory(BaseFactory["BaseDataset"]):
             stride = window_size
 
         dataset = cls.create(train_type, window_size, stride)
-        dataset.load(load_path, storage_type=storage_type)
+        dataset.load(load_path, storage_type=storage_type, **kwargs)
 
         return dataset
 
-    @classmethod
-    def available_types(cls) -> list:
-        """Return list of registered dataset type names."""
-        return cls.list_registered()
-
 
 @DatasetFactory.register("seq")
 class SEQDataset(BaseDataset):
     """Dataset for sequential next-token prediction training."""
 
-    def __init__(self, window_size: int, stride: int):
-        super().__init__(window_size, stride)
-
     @property
     def required_keys(self) -> List[str]:
         return ["sequence"]
@@ -218,12 +194,9 @@ class SEQDataset(BaseDataset):
 class SFTDataset(BaseDataset):
     """Dataset for supervised fine-tuning with loss masking."""
 
-    def __init__(self, window_size: int, stride: int):
-        super().__init__(window_size, stride)
-
     @property
     def required_keys(self) -> List[str]:
-        return ["sequence", "loss_mask"]
+        return ["sequence", "loss_mask", "position_ids"]
 
     def _fetch_data(self, begin_idx: int, end_idx: int, key: str) -> Tensor:
         return self.storage.fetch(begin_idx, end_idx, key)
@@ -231,24 +204,23 @@ class SFTDataset(BaseDataset):
     def __getitem__(self, index):
         begin_idx, end_idx = self.get_index(index)
 
-        x = self._fetch_data(begin_idx, end_idx, "sequence").to(dtype=torch.long)
-        y = self._fetch_data(begin_idx + 1, end_idx + 1, "sequence").to(
-            dtype=torch.long
-        )
-        loss_mask = self._fetch_data(begin_idx + 1, end_idx + 1, "loss_mask").to(
-            dtype=torch.bool
-        )
+        x = self._fetch_data(begin_idx, end_idx, "sequence")
+        y = self._fetch_data(begin_idx + 1, end_idx + 1, "sequence")
+        position_ids = self._fetch_data(begin_idx, end_idx, "position_ids")
+        loss_mask = self._fetch_data(begin_idx + 1, end_idx + 1, "loss_mask")
 
-        return {"input_ids": x, "target_ids": y, "loss_mask": loss_mask}
+        return {
+            "input_ids": x.to(dtype=torch.long),
+            "target_ids": y.to(dtype=torch.long),
+            "position_ids": position_ids.to(dtype=torch.long),
+            "loss_mask": loss_mask.to(dtype=torch.bool),
+        }
 
 
 @DatasetFactory.register("dpo")
 class DPODataset(BaseDataset):
     """Dataset for Direct Preference Optimization training."""
 
-    def __init__(self, window_size: int, stride: int):
-        super().__init__(window_size, stride)
-
     @property
     def required_keys(self) -> List[str]:
         return ["chosen", "rejected", "chosen_mask", "rejected_mask"]
@@ -280,9 +252,6 @@ class DPODataset(BaseDataset):
 class GRPODataset(BaseDataset):
     """Dataset for Group Relative Policy Optimization training."""
 
-    def __init__(self, window_size: int, stride: int):
-        super().__init__(window_size, stride)
-
     @property
     def required_keys(self) -> List[str]:
         return ["prompts", "responses", "masks", "rewards"]

astrai/dataset/sampler.py

@@ -74,6 +74,7 @@ class ResumableDistributedSampler(Sampler[int]):
 
         self.epoch += 1
         self._indices = None
+        self.iter = self.iter % self.num_samples_per_replica
 
     @property
     def _remaining(self):

astrai/dataset/storage.py

@@ -14,84 +14,31 @@ Key properties:
   - Explicit length: _length = min(total elements across keys), set at load,
                       __len__ returns O(1)
   - Zero-copy mmap:  MmapStore wraps np.memmap(mode="r"), all DataLoader
-                      workers share OS page-cache pages
+                       workers share OS page-cache pages
 """
 
 import bisect
+import glob
 import json
-import os
+import logging
 from abc import ABC, abstractmethod
 from pathlib import Path
 from typing import Dict, List, Union
 
-import h5py
-import numpy as np
 import torch
 from torch import Tensor
 
+from astrai.config.preprocess_config import PipelineConfig
 from astrai.factory import BaseFactory
+from astrai.preprocessing.builder import MaskBuilderFactory
+from astrai.preprocessing.position_id import PositionIdStrategyFactory
+from astrai.serialization import (
+    load_bin,
+    load_h5,
+)
+from astrai.tokenize import AutoTokenizer
 
-
-def save_h5(file_path: str, file_name: str, tensor_group: Dict[str, List[Tensor]]):
-    os.makedirs(file_path, exist_ok=True)
-    full_file_path = os.path.join(file_path, f"{file_name}.h5")
-    with h5py.File(full_file_path, "w") as f:
-        for key, tensors in tensor_group.items():
-            grp = f.create_group(key)
-            for idx, tensor in enumerate(tensors):
-                arr = tensor.cpu().numpy()
-                grp.create_dataset(f"data_{idx}", data=arr)
-
-
-def load_h5(file_path: str, share_memory=True) -> Dict[str, List[Tensor]]:
-    tensor_group: Dict[str, List[Tensor]] = {}
-
-    root_path = Path(file_path)
-    h5_files = list(root_path.rglob("*.h5")) + list(root_path.rglob("*.hdf5"))
-
-    for h5_file in h5_files:
-        with h5py.File(h5_file, "r") as f:
-            for key in f.keys():
-                grp = f[key]
-                dsets = []
-                for dset_name in grp.keys():
-                    dset = grp[dset_name]
-                    tensor = torch.from_numpy(dset[:])
-                    if share_memory:
-                        tensor = tensor.share_memory_()
-                    dsets.append(tensor)
-
-                if tensor_group.get(key) is None:
-                    tensor_group[key] = []
-                tensor_group[key].extend(dsets)
-
-    return tensor_group
-
-
-def save_bin(file_path: str, tensor_group: Dict[str, List[Tensor]]):
-    os.makedirs(file_path, exist_ok=True)
-    meta = {}
-    for key, tensors in tensor_group.items():
-        cat = torch.cat(tensors, dim=0)
-        meta[key] = {"shape": list(cat.shape), "dtype": str(cat.dtype).split(".")[-1]}
-        np.asarray(cat.cpu().numpy()).tofile(os.path.join(file_path, f"{key}.bin"))
-    with open(os.path.join(file_path, "meta.json"), "w") as f:
-        json.dump(meta, f)
-
-
-def load_bin(file_path: str) -> Dict[str, List[Tensor]]:
-    with open(os.path.join(file_path, "meta.json"), "r") as f:
-        meta = json.load(f)
-    segments: Dict[str, List[Tensor]] = {}
-    for key, info in meta.items():
-        arr = np.memmap(
-            os.path.join(file_path, f"{key}.bin"),
-            dtype=info["dtype"],
-            mode="r+",
-            shape=tuple(info["shape"]),
-        )
-        segments[key] = [torch.from_numpy(arr)]
-    return segments
+logger = logging.getLogger(__name__)
 
 
 def detect_format(load_path: str) -> str:
@@ -101,7 +48,7 @@ def detect_format(load_path: str) -> str:
         load_path: Directory or file path
 
     Returns:
-        Format string ("h5" or "bin")
+        Format string ("h5", "bin", or "jsonl")
 
     Raises:
         FileNotFoundError: If no supported data files are found
@@ -111,14 +58,29 @@ def detect_format(load_path: str) -> str:
         suffix = root.suffix.lower()
         if suffix in (".h5", ".hdf5"):
             return "h5"
+        if suffix == ".jsonl":
+            return "jsonl"
         raise ValueError(f"Unsupported file format: {suffix}")
 
-    h5_files = list(root.rglob("*.h5")) + list(root.rglob("*.hdf5"))
+    h5_files = [
+        Path(p)
+        for pattern in ("*.h5", "*.hdf5")
+        for p in glob.glob(str(root / "**" / pattern), recursive=True)
+    ]
     if h5_files:
         return "h5"
-    bin_files = list(root.rglob("*.bin"))
-    if bin_files and (root / "meta.json").exists():
-        return "bin"
+    bin_files = [Path(p) for p in glob.glob(str(root / "**" / "*.bin"), recursive=True)]
+    if bin_files:
+        has_meta = (root / "meta.json").exists() or len(
+            [Path(p) for p in glob.glob(str(root / "**" / "meta.json"), recursive=True)]
+        ) > 0
+        if has_meta:
+            return "bin"
+    jsonl_files = [
+        Path(p) for p in glob.glob(str(root / "**" / "*.jsonl"), recursive=True)
+    ]
+    if jsonl_files:
+        return "jsonl"
     raise FileNotFoundError(f"No supported data files found at {load_path}")
 
 
@@ -213,11 +175,6 @@ class StoreFactory(BaseFactory["Store"]):
             ...
     """
 
-    @classmethod
-    def _validate_component(cls, store_cls: type):
-        if not issubclass(store_cls, Store):
-            raise TypeError(f"{store_cls.__name__} must inherit from Store")
-
 
 @StoreFactory.register("h5")
 class H5Store(Store):
@@ -244,7 +201,112 @@ class MmapStore(Store):
 
     def load(self, path: str):
         self._mmap_refs = []
-        raw = load_bin(path)
-        self._normalize(raw)
+        root = Path(path)
+        all_raw: Dict[str, List[Tensor]] = {}
+        meta_paths = [
+            Path(p) for p in glob.glob(str(root / "**" / "meta.json"), recursive=True)
+        ]
+        for meta_path in meta_paths:
+            raw = load_bin(str(meta_path.parent))
+            for key, tensors in raw.items():
+                if key not in all_raw:
+                    all_raw[key] = []
+                all_raw[key].extend(tensors)
+        if not meta_paths:
+            raise FileNotFoundError(f"No meta.json found under {path}")
+        self._normalize(all_raw)
         for tensors in self._data.values():
             self._mmap_refs.extend(tensors)
+
+
+@StoreFactory.register("jsonl")
+class JsonlStore(Store):
+    """On-the-fly tokenization store for raw JSONL files.
+
+    A JSONL dataset directory contains ``*.jsonl`` files plus a
+    ``dataset_config.json`` file that follows the same schema as
+    :class:`PipelineConfig` with an additional ``tokenizer_path`` field.
+    Records are tokenized when the store is loaded and concatenated into
+    segmented tensors matching the key layout expected by the dataset
+    classes (``sequence``, ``loss_mask``, ``position_ids``, ...).
+    """
+
+    CONFIG_NAME = "dataset_config.json"
+
+    def load(self, path: str):
+        root = Path(path)
+        config_path = root / self.CONFIG_NAME
+        if not config_path.exists():
+            raise FileNotFoundError(
+                f"JSONL dataset config not found: {config_path}. "
+                f"Expected {self.CONFIG_NAME} alongside *.jsonl files."
+            )
+
+        with open(config_path, "r", encoding="utf-8") as f:
+            raw_config = json.load(f)
+
+        tokenizer_path = raw_config.pop("tokenizer_path", None)
+        if tokenizer_path is None:
+            raise ValueError(
+                f"JSONL dataset config must specify 'tokenizer_path': {config_path}"
+            )
+
+        self.config = PipelineConfig.from_dict(raw_config)
+        tokenizer = AutoTokenizer.from_pretrained(tokenizer_path)
+        mask_builder = MaskBuilderFactory.create("sectioned")
+        position_strategy = PositionIdStrategyFactory.create(
+            self.config.output.position_ids_mode
+        )
+
+        raw: Dict[str, List[Tensor]] = {}
+        doc_sequences: List[List[int]] = []
+
+        for jsonl_path in sorted(root.glob("*.jsonl")):
+            with open(jsonl_path, "r", encoding="utf-8") as f:
+                for line in f:
+                    line = line.strip()
+                    if not line:
+                        continue
+                    try:
+                        item = json.loads(line)
+                    except json.JSONDecodeError:
+                        logger.warning(
+                            "Failed to parse JSON line in %s, skipping", jsonl_path
+                        )
+                        continue
+
+                    result = mask_builder.build(item, self.config, tokenizer)
+                    if result is None:
+                        continue
+
+                    result.pop("domain", None)
+                    primary_ids = self._primary_ids(result)
+                    if not primary_ids:
+                        continue
+
+                    doc_sequences.append(primary_ids)
+                    for key, ids in result.items():
+                        if key not in raw:
+                            raw[key] = []
+                        raw[key].append(torch.tensor(ids, dtype=self._infer_dtype(ids)))
+
+        pos_ids = position_strategy.generate(doc_sequences)
+        if pos_ids:
+            raw["position_ids"] = [torch.tensor(pos_ids, dtype=torch.int32)]
+
+        self._normalize(raw)
+
+    @staticmethod
+    def _primary_ids(result: dict) -> List[int]:
+        """Return the first integer list in *result* as the primary id sequence."""
+        for val in result.values():
+            if isinstance(val, list) and val and isinstance(val[0], int):
+                return val
+        return []
+
+    @staticmethod
+    def _infer_dtype(ids: List) -> torch.dtype:
+        """Infer tensor dtype from the first element of a token/value list."""
+        if ids and isinstance(ids[0], float):
+            return torch.float32
+        return torch.int32

astrai/extension/__init__.py

@@ -0,0 +1,21 @@
+"""CUDA attention kernel wrappers with torch fallback.
+
+Public API:
+    - ``attn_decode`` — single-query decode attention
+    - ``attn_prefill`` — multi-query prefill attention
+    - ``attn_paged_decode`` — paged decode attention (direct page-table access)
+
+Each wrapper dispatches to its compiled CUDA kernel (``astrai.extension.attn_*``)
+when available, otherwise falls back to ``torch.nn.functional.scaled_dot_product_attention``.
+"""
+
+from astrai.extension.loader import KERNEL_NAMES, is_available
+from astrai.extension.ops import attn_decode, attn_paged_decode, attn_prefill
+
+__all__ = [
+    "attn_decode",
+    "attn_paged_decode",
+    "attn_prefill",
+    "is_available",
+    "KERNEL_NAMES",
+]

astrai/extension/loader.py

@@ -0,0 +1,36 @@
+"""Dynamic discovery and loading of compiled CUDA kernel modules.
+
+Each kernel is registered in ``csrc/build.py`` and built into a ``.so`` placed
+in this package directory. On import we try to load each one; kernels that
+failed to build (or are running on a CPU-only machine) are marked unavailable
+so the wrapper functions can fall back to ``torch`` SDPA.
+"""
+
+import importlib
+import logging
+
+logger = logging.getLogger(__name__)
+
+KERNEL_NAMES = ["attn_decode", "attn_prefill", "attn_paged_decode"]
+
+_available: dict[str, bool] = {}
+_modules: dict[str, object] = {}
+
+for _name in KERNEL_NAMES:
+    try:
+        _mod = importlib.import_module(f".{_name}", package=__package__)
+        _available[_name] = True
+        _modules[_name] = _mod
+    except ImportError:
+        _available[_name] = False
+        _modules[_name] = None
+
+
+def is_available(name: str) -> bool:
+    """Return ``True`` if the compiled kernel ``name`` was loaded."""
+    return _available.get(name, False)
+
+
+def get_module(name: str) -> object:
+    """Return the loaded kernel module for ``name``, or ``None`` if unavailable."""
+    return _modules.get(name)

astrai/extension/ops.py

@@ -0,0 +1,149 @@
+"""GQA attention wrapper functions — one entry point per compiled kernel.
+
+Each wrapper dispatches to its CUDA kernel (loaded in ``loader.py``) when
+available, otherwise falls back to ``torch`` SDPA.
+
+Add new kernel wrappers here; split into per-variant files only if this file
+grows large.
+"""
+
+import torch
+import torch.nn.functional as F
+
+from astrai.extension.loader import _available, _modules
+
+
+def _expand_kv_heads(
+    k: torch.Tensor, v: torch.Tensor, q_head: int
+) -> tuple[torch.Tensor, torch.Tensor]:
+    """Expand K/V heads to match Q heads for GQA fallback."""
+    kv_head = k.size(1)
+    if kv_head == q_head:
+        return k, v
+    group = q_head // kv_head
+    k = k.repeat_interleave(group, dim=1)
+    v = v.repeat_interleave(group, dim=1)
+    return k, v
+
+
+def _torch_fallback(
+    q: torch.Tensor,
+    k: torch.Tensor,
+    v: torch.Tensor,
+    mask: torch.Tensor | None,
+    is_causal: bool,
+    scale: float | None,
+) -> torch.Tensor:
+    """Reference attention via ``scaled_dot_product_attention``."""
+    k, v = _expand_kv_heads(k, v, q.size(1))
+    attn_mask = mask[:, None, None, :] if mask is not None else None
+    return F.scaled_dot_product_attention(
+        q, k, v, attn_mask=attn_mask, is_causal=is_causal and mask is None, scale=scale
+    )
+
+
+def _gather_kv_from_pages(
+    page_table: torch.Tensor,
+    k_cache: torch.Tensor,
+    v_cache: torch.Tensor,
+    page_size: int,
+    kv_len: int,
+) -> tuple[torch.Tensor, torch.Tensor]:
+    """Gather contiguous K/V from paged cache for torch SDPA fallback.
+
+    Shapes:
+        page_table : [batch, max_pages] (int64)
+        k_cache    : [n_pages, page_size, n_kv_heads, head_dim]
+        v_cache    : same as k_cache
+    Returns:
+        k, v : [batch, n_kv_heads, kv_len, head_dim]
+    """
+    batch, max_pages = page_table.shape
+    n_pages, ps, n_kv_heads, head_dim = k_cache.shape
+    if ps != page_size:
+        raise ValueError(f"k_cache page_size mismatch: {ps} vs {page_size}")
+
+    k = k_cache.new_empty(batch, n_kv_heads, kv_len, head_dim)
+    v = v_cache.new_empty(batch, n_kv_heads, kv_len, head_dim)
+
+    for b in range(batch):
+        for pos in range(kv_len):
+            log_pg = pos // page_size
+            pg_off = pos % page_size
+            phys = int(page_table[b, log_pg].item())
+            k[b, :, pos, :] = k_cache[phys, pg_off, :, :]
+            v[b, :, pos, :] = v_cache[phys, pg_off, :, :]
+    return k, v
+
+
+def attn_decode(
+    q: torch.Tensor,
+    k: torch.Tensor,
+    v: torch.Tensor,
+    mask: torch.Tensor | None = None,
+    is_causal: bool = False,
+    causal_offset: int = 0,
+    scale: float | None = None,
+) -> torch.Tensor:
+    if _available["attn_decode"]:
+        return _modules["attn_decode"].attn_decode(
+            q,
+            k,
+            v,
+            mask=mask,
+            is_causal=is_causal,
+            causal_offset=causal_offset,
+            scale=scale,
+        )
+    return _torch_fallback(q, k, v, mask, is_causal, scale)
+
+
+def attn_prefill(
+    q: torch.Tensor,
+    k: torch.Tensor,
+    v: torch.Tensor,
+    mask: torch.Tensor | None = None,
+    is_causal: bool = False,
+    causal_offset: int = 0,
+    scale: float | None = None,
+) -> torch.Tensor:
+    if _available["attn_prefill"]:
+        return _modules["attn_prefill"].attn_prefill(
+            q,
+            k,
+            v,
+            mask=mask,
+            is_causal=is_causal,
+            causal_offset=causal_offset,
+            scale=scale,
+        )
+    return _torch_fallback(q, k, v, mask, is_causal, scale)
+
+
+def attn_paged_decode(
+    q: torch.Tensor,
+    page_table: torch.Tensor,
+    k_cache: torch.Tensor,
+    v_cache: torch.Tensor,
+    page_size: int,
+    kv_len: int,
+    mask: torch.Tensor | None = None,
+    is_causal: bool = False,
+    causal_offset: int = 0,
+    scale: float | None = None,
+) -> torch.Tensor:
+    if _available["attn_paged_decode"]:
+        return _modules["attn_paged_decode"].attn_paged_decode(
+            q,
+            page_table,
+            k_cache,
+            v_cache,
+            page_size,
+            kv_len,
+            mask=mask,
+            is_causal=is_causal,
+            causal_offset=causal_offset,
+            scale=scale,
+        )
+    k, v = _gather_kv_from_pages(page_table, k_cache, v_cache, page_size, kv_len)
+    return _torch_fallback(q, k, v, mask, is_causal, scale)

astrai/factory.py

@@ -1,149 +1,103 @@
-"""Base factory class for extensible component registration."""
+"""Base factory with decorator-based registration and kwarg-filtered instantiation."""
 
 import inspect
+import sys
 from abc import ABC
-from typing import Callable, Dict, Generic, List, Optional, Tuple, Type, TypeVar
+from typing import (
+    Callable,
+    Dict,
+    ForwardRef,
+    Generic,
+    List,
+    Optional,
+    Type,
+    TypeVar,
+    Union,
+)
+from typing import get_args as _get_args
+from typing import get_origin as _get_origin
 
 T = TypeVar("T")
 
 
-class Registry:
-    """Flexible registry for component classes with category and priority support.
+def _resolve_type(
+    arg: Union[Type, str, ForwardRef], factory_cls: type
+) -> Optional[Type]:
+    """Resolve a generic type-arg (str forward-ref, ForwardRef, or class)."""
+    if not isinstance(arg, (str, ForwardRef)):
+        return arg
 
-    This registry stores component classes with optional metadata (category, priority).
-    It provides methods for registration, retrieval, and listing with filtering.
-    """
+    name = arg if isinstance(arg, str) else arg.__forward_arg__
+    if name == factory_cls.__name__:
+        return factory_cls
 
-    def __init__(self):
-        self._entries = {}  # name -> (component_cls, category, priority)
+    mod = sys.modules.get(factory_cls.__module__)
+    if mod is None:
+        return None
+    ns = vars(mod)
 
-    def register(
-        self,
-        name: str,
-        component_cls: Type,
-        category: Optional[str] = None,
-        priority: int = 0,
-    ):
-        """Register a component class with optional category and priority."""
-        if name in self._entries:
-            raise ValueError(f"Component '{name}' is already registered")
-        self._entries[name] = (component_cls, category, priority)
+    if isinstance(arg, ForwardRef):
+        return arg._evaluate(ns, None, recursive_guard=frozenset())
 
-    def get(self, name: str) -> Type:
-        """Get component class by name."""
-        if name not in self._entries:
-            raise KeyError(f"Component '{name}' not found in registry")
-        return self._entries[name][0]
-
-    def get_with_metadata(self, name: str) -> Tuple[Type, Optional[str], int]:
-        """Get component class with its metadata."""
-        entry = self._entries.get(name)
-        if entry is None:
-            raise KeyError(f"Component '{name}' not found in registry")
-        return entry
-
-    def contains(self, name: str) -> bool:
-        """Check if a name is registered."""
-        return name in self._entries
-
-    def list_names(self) -> List[str]:
-        """Return list of registered component names."""
-        return sorted(self._entries.keys())
-
-    def list_by_category(self, category: str) -> List[str]:
-        """Return names of components belonging to a specific category."""
-        return sorted(
-            name for name, (_, cat, _) in self._entries.items() if cat == category
-        )
-
-    def list_by_priority(self, reverse: bool = False) -> List[str]:
-        """Return names sorted by priority (default ascending)."""
-        return sorted(
-            self._entries.keys(),
-            key=lambda name: self._entries[name][2],
-            reverse=reverse,
-        )
-
-    def entries(self) -> Dict[str, Tuple[Type, Optional[str], int]]:
-        """Return raw entries dictionary."""
-        return self._entries.copy()
+    return ns.get(name)
 
 
 class BaseFactory(ABC, Generic[T]):
-    """Generic factory class for component registration and creation.
+    """Generic factory with decorator-based component registration.
 
-    This base class provides a decorator-based registration pattern
-    for creating extensible component factories.
-
-    Example usage:
-        class MyFactory(BaseFactory[MyBaseClass]):
+        class MyFactory(BaseFactory[MyBase]):
             pass
 
         @MyFactory.register("custom")
-        class CustomComponent(MyBaseClass):
+        class CustomComponent(MyBase):
             ...
 
-        component = MyFactory.create("custom", *args, **kwargs)
+        obj = MyFactory.create("custom", *args, **kwargs)
+
+    ``create()`` filters kwargs to match the component's ``__init__``
+    signature so components don't need ``**kwargs`` just to absorb
+    unrelated parameters.
     """
 
-    _registry: Registry
+    _entries: Dict[str, Type[T]]
 
     def __init_subclass__(cls, **kwargs):
         super().__init_subclass__(**kwargs)
-        cls._registry = Registry()
+        for orig_base in getattr(cls, "__orig_bases__", ()):
+            if _get_origin(orig_base) is BaseFactory:
+                (arg,) = _get_args(orig_base)
+                cls._entries = {}
+                cls._component_base = _resolve_type(arg, cls)
+                return
 
     @classmethod
-    def register(
-        cls, name: str, category: Optional[str] = None, priority: int = 0
-    ) -> Callable[[Type[T]], Type[T]]:
-        """Decorator to register a component class with optional category and priority.
+    def register(cls, name: str) -> Callable[[Type[T]], Type[T]]:
+        """Decorator to register a component class.
 
-        Args:
-            name: Registration name for the component
-            category: Optional category for grouping components
-            priority: Priority for ordering (default 0)
-
-        Returns:
-            Decorator function that registers the component class
-
-        Raises:
-            TypeError: If the decorated class doesn't inherit from the base type
+        Validates that the decorated class inherits from the generic
+        type parameter ``T`` declared on the factory.
         """
 
         def decorator(component_cls: Type[T]) -> Type[T]:
             cls._validate_component(component_cls)
-            cls._registry.register(
-                name, component_cls, category=category, priority=priority
-            )
+            if name in cls._entries:
+                raise ValueError(f"Component '{name}' is already registered")
+            cls._entries[name] = component_cls
             return component_cls
 
         return decorator
 
     @classmethod
     def create(cls, name: str, *args, **kwargs) -> T:
-        """Create a component instance by name.
-
-        Filters kwargs to match the component's __init__ signature,
-        so components don't need to declare **kwargs just to absorb
-        parameters meant for other components.
-
-        Args:
-            name: Registered name of the component
-            *args: Positional arguments passed to component constructor
-            **kwargs: Keyword arguments passed to component constructor
-
-        Returns:
-            Component instance
-
-        Raises:
-            ValueError: If the component name is not registered
+        """Create a component instance by name, filtering kwargs to match
+        the component's ``__init__`` signature.
         """
-        if not cls._registry.contains(name):
+        entry = cls._entries.get(name)
+        if entry is None:
             raise ValueError(
-                f"Unknown component: '{name}'. "
-                f"Supported types: {sorted(cls._registry.list_names())}"
+                f"Unknown component: '{name}'. Supported types: {sorted(cls._entries)}"
             )
-        component_cls = cls._registry.get(name)
+        component_cls = entry
         sig = inspect.signature(component_cls.__init__)
         has_var_kwargs = any(
             p.kind == inspect.Parameter.VAR_KEYWORD for p in sig.parameters.values()
@@ -159,68 +113,32 @@ class BaseFactory(ABC, Generic[T]):
 
     @classmethod
     def _validate_component(cls, component_cls: Type[T]):
-        """Validate that the component class is valid for this factory.
+        """Validate the decorated class inherits from the factory's base type.
 
-        Override this method in subclasses to add custom validation.
-
-        Args:
-            component_cls: Component class to validate
-
-        Raises:
-            TypeError: If the component class is invalid
+        Override for custom validation beyond ``issubclass``.
         """
-        pass
+        base = cls._component_base
+        if base is not None and not issubclass(component_cls, base):
+            raise TypeError(
+                f"{component_cls.__name__} must inherit from {base.__name__}"
+            )
 
     @classmethod
     def get_component_class(cls, name: str) -> Type[T]:
-        """Get the registered component class by name without instantiating it.
-
-        Args:
-            name: Registered name of the component
-
-        Returns:
-            The component class itself
-
-        Raises:
-            ValueError: If the component name is not registered
-        """
-        if not cls._registry.contains(name):
+        """Get the registered component class without instantiating it."""
+        entry = cls._entries.get(name)
+        if entry is None:
             raise ValueError(
-                f"Unknown component: '{name}'. "
-                f"Supported types: {sorted(cls._registry.list_names())}"
+                f"Unknown component: '{name}'. Supported types: {sorted(cls._entries)}"
             )
-        return cls._registry.get(name)
+        return entry
 
     @classmethod
-    def list_registered(cls) -> list:
-        """List all registered component names.
-
-        Returns:
-            List of registered component names
-        """
-        return cls._registry.list_names()
+    def list_registered(cls) -> List[str]:
+        """List all registered component names."""
+        return sorted(cls._entries)
 
     @classmethod
     def is_registered(cls, name: str) -> bool:
-        """Check if a component name is registered.
-
-        Args:
-            name: Component name to check
-
-        Returns:
-            True if registered, False otherwise
-        """
-        return cls._registry.contains(name)
-
-    @classmethod
-    def list_by_category(cls, category: str) -> List[str]:
-        """List registered component names in a category."""
-        return cls._registry.list_by_category(category)
-
-    @classmethod
-    def list_by_priority(cls, reverse: bool = False) -> List[str]:
-        """List registered component names sorted by priority."""
-        return cls._registry.list_by_priority(reverse)
-
-
-__all__ = ["Registry", "BaseFactory"]
+        """Check if a component name is registered."""
+        return name in cls._entries

astrai/inference/__init__.py

@@ -11,13 +11,18 @@ Layers:
 
 from astrai.inference.api import (
     AnthropicMessage,
+    BaseToolParser,
     ChatCompletionRequest,
     ChatMessage,
+    FunctionDef,
     GenContext,
     MessagesRequest,
     ProtocolHandler,
+    SimpleJsonToolParser,
     StopChecker,
-    app,
+    ToolDef,
+    ToolParserFactory,
+    get_app,
     run_server,
 )
 from astrai.inference.api.anthropic import AnthropicResponseBuilder
@@ -25,10 +30,14 @@ from astrai.inference.api.openai import OpenAIResponseBuilder
 from astrai.inference.core import (
     STOP,
     Allocator,
+    CacheView,
+    ContiguousCache,
+    ContiguousCacheView,
     Executor,
     InferenceScheduler,
     KVCache,
-    KvcacheView,
+    PageCache,
+    PageCacheView,
     PagePool,
     PrefixCache,
     Storage,
@@ -58,8 +67,12 @@ __all__ = [
     "TaskManager",
     "TaskStatus",
     "Allocator",
+    "CacheView",
     "KVCache",
-    "KvcacheView",
+    "ContiguousCache",
+    "ContiguousCacheView",
+    "PageCache",
+    "PageCacheView",
     "PagePool",
     "PrefixCache",
     "Storage",
@@ -74,12 +87,17 @@ __all__ = [
     "ProtocolHandler",
     "StopChecker",
     "GenContext",
+    "BaseToolParser",
+    "SimpleJsonToolParser",
+    "ToolParserFactory",
     "OpenAIResponseBuilder",
     "AnthropicResponseBuilder",
     "ChatMessage",
     "ChatCompletionRequest",
+    "FunctionDef",
+    "ToolDef",
     "AnthropicMessage",
     "MessagesRequest",
-    "app",
+    "get_app",
     "run_server",
 ]

astrai/inference/api/__init__.py

@@ -1,23 +1,39 @@
-"""Inference API: protocol handler, stop checker, and FastAPI server."""
+"""Inference API: protocol handler, stop checker, tool parsers, and FastAPI server.
+
+``app`` is no longer a module-level global. Use :func:`get_app` to access the
+lazy singleton FastAPI instance.
+"""
 
 from astrai.inference.api.protocol import GenContext, ProtocolHandler, StopChecker
 from astrai.inference.api.server import (
     AnthropicMessage,
     ChatCompletionRequest,
     ChatMessage,
+    FunctionDef,
     MessagesRequest,
-    app,
+    ToolDef,
+    get_app,
     run_server,
 )
+from astrai.inference.api.tool_parser import (
+    BaseToolParser,
+    SimpleJsonToolParser,
+    ToolParserFactory,
+)
 
 __all__ = [
     "ProtocolHandler",
     "StopChecker",
     "GenContext",
+    "BaseToolParser",
+    "SimpleJsonToolParser",
+    "ToolParserFactory",
     "AnthropicMessage",
     "ChatCompletionRequest",
     "ChatMessage",
+    "FunctionDef",
+    "ToolDef",
     "MessagesRequest",
-    "app",
+    "get_app",
     "run_server",
 ]

astrai/inference/api/anthropic.py

@@ -1,5 +1,6 @@
 """Anthropic message completion response builder."""
 
+import time
 import uuid
 from typing import Any, Dict, List, Tuple, Union
 
@@ -39,9 +40,8 @@ class AnthropicResponseBuilder(ResponseBuilder):
         prompt = engine.tokenizer.apply_chat_template(messages, tokenize=False)
         ctx = GenContext(
             resp_id=f"msg_{uuid.uuid4().hex[:24]}",
-            created=0,
+            created=int(time.time()),
             model=request.model,
-            prompt_tokens=0,
         )
         stop_sequences = getattr(request, "stop_sequences", None) or []
         return prompt, ctx, stop_sequences
@@ -72,15 +72,17 @@ class AnthropicResponseBuilder(ResponseBuilder):
             ),
         ]
 
-    def format_chunk(self, token: str) -> str:
-        return sse_event(
-            {
-                "type": "content_block_delta",
-                "index": 0,
-                "delta": {"type": "text_delta", "text": token},
-            },
-            event="content_block_delta",
-        )
+    def format_chunk(self, token: str, **kwargs) -> List[str]:
+        return [
+            sse_event(
+                {
+                    "type": "content_block_delta",
+                    "index": 0,
+                    "delta": {"type": "text_delta", "text": token},
+                },
+                event="content_block_delta",
+            )
+        ]
 
     def format_stream_end(self, ctx: GenContext, stop: StopInfo) -> List[str]:
         events: List[str] = []

astrai/inference/api/openai.py

@@ -1,7 +1,9 @@
 """OpenAI chat completion response builder."""
 
+import logging
+import time
 import uuid
-from typing import Any, Dict, List, Tuple
+from typing import Any, Dict, List, Optional, Tuple, Union
 
 from pydantic import BaseModel
 
@@ -11,24 +13,79 @@ from astrai.inference.api.protocol import (
     StopInfo,
     sse_event,
 )
+from astrai.inference.api.tool_parser import BaseToolParser, ToolParserFactory
 from astrai.inference.engine import InferenceEngine
 
+logger = logging.getLogger(__name__)
+
+_UNSUPPORTED_PARAMS = (
+    "n",
+    "presence_penalty",
+    "frequency_penalty",
+    "logit_bias",
+    "user",
+)
+
+
+def _resolve_tool_choice(
+    request: BaseModel,
+) -> Union[str, Dict[str, Any]]:
+    tc = getattr(request, "tool_choice", None)
+    if tc is None:
+        return "auto"
+    if isinstance(tc, str):
+        return tc
+    if isinstance(tc, dict):
+        return tc
+    return "auto"
+
+
+def _resolve_tools(request: BaseModel) -> Optional[List[Dict[str, Any]]]:
+    raw = getattr(request, "tools", None)
+    if not raw:
+        return None
+    if isinstance(raw, list):
+        return [t.model_dump() if hasattr(t, "model_dump") else t for t in raw]
+    return None
+
 
 class OpenAIResponseBuilder(ResponseBuilder):
     def prepare(
         self, request: BaseModel, engine: InferenceEngine
     ) -> Tuple[str, GenContext, List[str]]:
         messages = [{"role": m.role, "content": m.content} for m in request.messages]
-        prompt = engine.tokenizer.apply_chat_template(messages, tokenize=False)
+        tools = _resolve_tools(request)
+        prompt = engine.tokenizer.apply_chat_template(
+            messages, tokenize=False, tools=tools or []
+        )
 
         self._resp_id = f"chatcmpl-{uuid.uuid4().hex[:12]}"
         self._model = request.model
 
+        for param in _UNSUPPORTED_PARAMS:
+            value = getattr(request, param, None)
+            fields = getattr(type(request), "model_fields", {})
+            default = fields[param].default if param in fields else None
+            if value is not None and value != default:
+                logger.warning(
+                    "ChatCompletionRequest param '%s'=%r is not supported"
+                    " and will be ignored",
+                    param,
+                    value,
+                )
+
+        self._parser: Optional[BaseToolParser] = None
+        if tools:
+            tool_choice = _resolve_tool_choice(request)
+            self._parser = ToolParserFactory.create(
+                "simple_json", tools=tools, tool_choice=tool_choice
+            )
+        self._content_started = False
+
         ctx = GenContext(
             resp_id=self._resp_id,
-            created=0,
+            created=int(time.time()),
             model=self._model,
-            prompt_tokens=0,
         )
         stop = request.stop
         stop_sequences = (
@@ -55,7 +112,82 @@ class OpenAIResponseBuilder(ResponseBuilder):
             )
         ]
 
-    def format_chunk(self, token: str) -> str:
+    def format_chunk(self, token: str, **kwargs) -> List[str]:
+        body = kwargs.get("body", "")
+        if self._parser is not None:
+            return self._format_tool_chunk(body, **kwargs)
+
+        return [
+            sse_event(
+                {
+                    "id": self._resp_id,
+                    "object": "chat.completion.chunk",
+                    "created": 0,
+                    "model": self._model,
+                    "choices": [
+                        {
+                            "index": 0,
+                            "delta": {"content": token},
+                            "finish_reason": None,
+                        }
+                    ],
+                }
+            )
+        ]
+
+    def _format_tool_chunk(self, body: str, **kwargs) -> List[str]:
+        deltas = self._parser.feed(
+            body,
+            current_token_ids=kwargs.get("current_token_ids"),
+            delta_token_ids=kwargs.get("delta_token_ids"),
+        )
+        events: List[str] = []
+        for d in deltas:
+            if "content" in d:
+                if not self._content_started:
+                    events.append(self._role_chunk())
+                    self._content_started = True
+                events.append(
+                    sse_event(
+                        {
+                            "id": self._resp_id,
+                            "object": "chat.completion.chunk",
+                            "created": 0,
+                            "model": self._model,
+                            "choices": [
+                                {
+                                    "index": 0,
+                                    "delta": {"content": d["content"]},
+                                    "finish_reason": None,
+                                }
+                            ],
+                        }
+                    )
+                )
+            elif "tool_calls" in d:
+                if not self._content_started:
+                    events.append(self._role_chunk())
+                    self._content_started = True
+                events.append(
+                    sse_event(
+                        {
+                            "id": self._resp_id,
+                            "object": "chat.completion.chunk",
+                            "created": 0,
+                            "model": self._model,
+                            "choices": [
+                                {
+                                    "index": 0,
+                                    "delta": {"tool_calls": d["tool_calls"]},
+                                    "finish_reason": None,
+                                }
+                            ],
+                        }
+                    )
+                )
+        return events
+
+    def _role_chunk(self) -> str:
         return sse_event(
             {
                 "id": self._resp_id,
@@ -63,12 +195,19 @@ class OpenAIResponseBuilder(ResponseBuilder):
                 "created": 0,
                 "model": self._model,
                 "choices": [
-                    {"index": 0, "delta": {"content": token}, "finish_reason": None}
+                    {
+                        "index": 0,
+                        "delta": {"role": "assistant"},
+                        "finish_reason": None,
+                    }
                 ],
             }
         )
 
     def format_stream_end(self, ctx: GenContext, stop: StopInfo) -> List[str]:
+        finish_reason = "stop"
+        if self._parser is not None and self._parser.has_tool_calls:
+            finish_reason = "tool_calls"
         return [
             sse_event(
                 {
@@ -76,7 +215,9 @@ class OpenAIResponseBuilder(ResponseBuilder):
                     "object": "chat.completion.chunk",
                     "created": ctx.created,
                     "model": self._model,
-                    "choices": [{"index": 0, "delta": {}, "finish_reason": "stop"}],
+                    "choices": [
+                        {"index": 0, "delta": {}, "finish_reason": finish_reason}
+                    ],
                 }
             ),
             sse_event(
@@ -91,6 +232,32 @@ class OpenAIResponseBuilder(ResponseBuilder):
     def format_response(
         self, ctx: GenContext, content: str, stop: StopInfo
     ) -> Dict[str, Any]:
+        if self._parser is not None:
+            parsed = self._parser.parse_complete(content)
+            if parsed and parsed.get("tool_calls"):
+                return {
+                    "id": self._resp_id,
+                    "object": "chat.completion",
+                    "created": ctx.created,
+                    "model": self._model,
+                    "choices": [
+                        {
+                            "index": 0,
+                            "message": {
+                                "role": "assistant",
+                                "content": parsed.get("content"),
+                                "tool_calls": parsed["tool_calls"],
+                            },
+                            "finish_reason": "tool_calls",
+                        }
+                    ],
+                    "usage": {
+                        "prompt_tokens": ctx.prompt_tokens,
+                        "completion_tokens": ctx.completion_tokens,
+                        "total_tokens": ctx.prompt_tokens + ctx.completion_tokens,
+                    },
+                }
+
         return {
             "id": self._resp_id,
             "object": "chat.completion",

astrai/inference/api/protocol.py

@@ -35,7 +35,7 @@ class GenContext:
     resp_id: str
     created: int
     model: str
-    prompt_tokens: int
+    prompt_tokens: int = 0
     completion_tokens: int = 0
 
 
@@ -64,7 +64,7 @@ class StopChecker:
 class ResponseBuilder(ABC):
     """Interface for protocol-specific response formatting.
 
-    A new protocol requires one concrete builder implementing 6 methods.
+    A new protocol requires one concrete builder implementing 5 methods.
     """
 
     @abstractmethod
@@ -78,8 +78,15 @@ class ResponseBuilder(ABC):
         """SSE events that open the stream."""
 
     @abstractmethod
-    def format_chunk(self, token: str) -> str:
-        """SSE event for a single generated token."""
+    def format_chunk(self, token: str, **kwargs) -> List[str]:
+        """SSE events for a single generated token.
+
+        ``body`` (the full accumulated text so far) is always provided
+        as a keyword argument. Additional keyword arguments such as
+        ``current_token_ids`` and ``delta_token_ids`` may be included
+        for tool parsers that need token-level information.
+        Returns a list of SSE event strings (may be empty).
+        """
 
     @abstractmethod
     def format_stream_end(self, ctx: GenContext, stop: StopInfo) -> List[str]:
@@ -137,15 +144,25 @@ class ProtocolHandler:
             body = ""
             yielded = ""
             matched = None
+            token_ids: List[int] = []
             async for token in agen:
-                ctx.completion_tokens += 1
                 body += token
 
+                new_ids = self.engine.tokenizer.encode(token)
+                token_ids.extend(new_ids)
+
                 matched = checker.check(body)
                 if matched:
                     break
 
-                yield self.builder.format_chunk(token)
+                ctx.completion_tokens += 1
+                for event in self.builder.format_chunk(
+                    token,
+                    body=body,
+                    current_token_ids=token_ids,
+                    delta_token_ids=new_ids,
+                ):
+                    yield event
                 yielded += token
 
             stop = StopInfo(matched=matched, body=body, yielded=yielded)
@@ -168,7 +185,6 @@ class ProtocolHandler:
         matched = None
 
         async for token in agen:
-            ctx.completion_tokens += 1
             chunks.append(token)
             body += token
 
@@ -176,6 +192,8 @@ class ProtocolHandler:
             if matched:
                 break
 
+            ctx.completion_tokens += 1
+
         content = "".join(chunks)
         stop = StopInfo(matched=matched, body=body)
         return self.builder.format_response(ctx, content, stop)

astrai/inference/api/server.py

@@ -3,6 +3,9 @@ OpenAI / Anthropic-compatible chat completion server backed by continuous-batchi
 
 Protocol-specific formatting is delegated to ``astrai.inference.protocol``.
 This module owns the FastAPI app, request/response schemas, and dependency wiring.
+
+``app`` is lazily constructed — importing this module does NOT create a FastAPI instance.
+Use :func:`get_app` to access the singleton.
 """
 
 import logging
@@ -12,7 +15,7 @@ from typing import Any, Dict, List, Optional, Union
 
 import torch
 import uvicorn
-from fastapi import FastAPI, HTTPException
+from fastapi import APIRouter, FastAPI, HTTPException
 from pydantic import BaseModel, Field
 
 from astrai.inference.api.anthropic import AnthropicResponseBuilder
@@ -24,12 +27,25 @@ from astrai.tokenize import AutoTokenizer
 
 logger = logging.getLogger(__name__)
 
-_project_root = Path(__file__).parent.parent.parent
+_app_instance: Optional[FastAPI] = None
 
 
 class ChatMessage(BaseModel):
     role: str
-    content: str
+    content: Optional[str] = None
+    tool_calls: Optional[List[Dict[str, Any]]] = None
+    tool_call_id: Optional[str] = None
+
+
+class FunctionDef(BaseModel):
+    name: str
+    description: Optional[str] = None
+    parameters: Optional[Dict[str, Any]] = None
+
+
+class ToolDef(BaseModel):
+    type: str = "function"
+    function: FunctionDef
 
 
 class ChatCompletionRequest(BaseModel):
@@ -48,6 +64,8 @@ class ChatCompletionRequest(BaseModel):
     frequency_penalty: Optional[float] = Field(default=0.0, ge=-2.0, le=2.0)
     logit_bias: Optional[Dict[int, float]] = None
     user: Optional[str] = None
+    tools: Optional[List[ToolDef]] = None
+    tool_choice: Optional[Union[str, Dict[str, Any]]] = "auto"
 
 
 class AnthropicMessage(BaseModel):
@@ -84,17 +102,15 @@ async def lifespan(app: FastAPI):
         logger.info("Inference engine shutdown complete")
 
 
-app = FastAPI(title="AstrAI Inference Server", version="0.2.0", lifespan=lifespan)
+router = APIRouter()
 
 
 def _create_engine(
-    param_path: Optional[Path] = None,
+    param_path: Path,
     device: str = "cuda",
     dtype: torch.dtype = torch.bfloat16,
     max_batch_size: int = 16,
 ) -> InferenceEngine:
-    if param_path is None:
-        param_path = _project_root / "params"
     if not param_path.exists():
         raise FileNotFoundError(f"Parameter directory not found: {param_path}")
 
@@ -112,34 +128,50 @@ def _create_engine(
     return engine
 
 
+def get_app() -> FastAPI:
+    """Return the singleton FastAPI instance (lazily created on first call)."""
+    global _app_instance
+    if _app_instance is None:
+        _app_instance = FastAPI(
+            title="AstrAI Inference Server",
+            version="0.2.0",
+            lifespan=lifespan,
+        )
+        _app_instance.include_router(router)
+        _app_instance.state.server_config = {}
+        _app_instance.state.engine = None
+    return _app_instance
+
+
 def _get_engine() -> InferenceEngine:
-    engine = app.state.engine
+    engine = get_app().state.engine
     if engine is None:
         raise HTTPException(status_code=503, detail="Engine not initialized")
     return engine
 
 
-@app.get("/health")
+@router.get("/health")
 async def health():
+    app = get_app()
     return {
         "status": "ok",
         "model_loaded": app.state.engine is not None,
     }
 
 
-@app.get("/stats")
+@router.get("/stats")
 async def get_stats():
     return _get_engine().get_stats()
 
 
-@app.post("/v1/chat/completions")
+@router.post("/v1/chat/completions")
 async def chat_completion(request: ChatCompletionRequest):
     engine = _get_engine()
     handler = ProtocolHandler(request, engine, OpenAIResponseBuilder())
     return await handler.handle()
 
 
-@app.post("/v1/messages")
+@router.post("/v1/messages")
 async def create_message(request: MessagesRequest):
     engine = _get_engine()
     handler = ProtocolHandler(request, engine, AnthropicResponseBuilder())
@@ -147,14 +179,15 @@ async def create_message(request: MessagesRequest):
 
 
 def run_server(
+    param_path: Path,
     host: str = "0.0.0.0",
     port: int = 8000,
     reload: bool = False,
     device: str = "cuda",
     dtype: torch.dtype = torch.bfloat16,
-    param_path: Optional[Path] = None,
     max_batch_size: int = 16,
 ):
+    app = get_app()
     app.state.server_config = {
         "device": device,
         "dtype": dtype,

astrai/inference/api/tool_parser.py

@@ -0,0 +1,325 @@
+"""Tool call parsers for extracting structured tool calls from model output.
+
+Patterned after vLLM's ToolParser abstraction. Each parser knows how to
+detect and incrementally extract tool calls from raw generated text.
+
+Subclasses may optionally consume ``token_ids`` for token-level parsing
+(e.g. Harmony / VLM-style parsers).
+"""
+
+import re
+import uuid
+from abc import ABC, abstractmethod
+from typing import Dict, List, Optional
+
+from astrai.factory import BaseFactory
+
+
+class BaseToolParser(ABC):
+    """Abstract tool call parser — one instance per request.
+
+    Maintains streaming state internally so that each call to :meth:`feed`
+    can diff against previously emitted content.
+
+    Parameters
+    ----------
+    tools : list of dict, optional
+        Tool definitions from the request.
+    tool_choice : str
+        ``"auto"`` / ``"required"`` / ``"none"`` or a named tool choice
+        dict.
+    """
+
+    def __init__(self, tools: Optional[List[Dict]] = None, tool_choice: str = "auto"):
+        self.tools = tools or []
+        self.tool_choice = tool_choice
+
+    @abstractmethod
+    def feed(
+        self,
+        body: str,
+        current_token_ids: Optional[List[int]] = None,
+        delta_token_ids: Optional[List[int]] = None,
+    ) -> List[Dict]:
+        """Feed the *full* accumulated text each step.
+
+        Returns a list of delta dicts to emit. Each delta is one of:
+
+        - ``{"content": "text"}``      — plain text delta
+        - ``{"tool_calls": [...]}``    — tool-call delta (OpenAI format)
+
+        Returns an empty list when nothing new should be emitted.
+
+        Parameters
+        ----------
+        body : str
+            The complete accumulated generated text so far.
+        current_token_ids : list of int, optional
+            All token IDs decoded into *body* (cumulative).
+        delta_token_ids : list of int, optional
+            Only the token IDs for this chunk.
+        """
+
+    @abstractmethod
+    def parse_complete(self, body: str) -> Optional[Dict]:
+        """Parse the *complete* generated text after generation ends.
+
+        Returns ``None`` when no tool calls were found, otherwise a dict
+        with ``content`` (str or None) and ``tool_calls`` (list of dicts).
+        """
+
+    @property
+    @abstractmethod
+    def has_tool_calls(self) -> bool:
+        """True if the parser detected at least one tool call in the stream."""
+
+
+class ToolParserFactory(BaseFactory["BaseToolParser"]):
+    pass
+
+
+_TOOL_CALL_HEAD_RE = re.compile(r'\{\s*"name"\s*:')
+
+
+def _scan_json(text: str, start: int = 0):
+    """Scan for a complete JSON object starting at *start*.
+
+    Returns ``(end, complete)`` where *end* is one-past the closing
+    brace (or ``len(text)`` if unclosed), and *complete* is a bool.
+    """
+    depth = 0
+    in_string = False
+    escape = False
+    for i in range(start, len(text)):
+        c = text[i]
+        if escape:
+            escape = False
+            continue
+        if c == "\\":
+            escape = True
+            continue
+        if c == '"':
+            in_string = not in_string
+            continue
+        if in_string:
+            continue
+        if c == "{":
+            depth += 1
+        elif c == "}":
+            depth -= 1
+            if depth == 0:
+                return i + 1, True
+    return len(text), False
+
+
+def _parse_tool_call_json(json_str: str, complete: bool):
+    """Extract *name* and *arguments* from a tool-call JSON string.
+
+    Returns ``(name, args, valid)``.
+    """
+    name_match = re.search(r'"name"\s*:\s*"([^"]*)"', json_str)
+    if not name_match:
+        return None, "", False
+    name = name_match.group(1)
+
+    args_match = re.search(r'"arguments"\s*:\s*(.*)', json_str, re.DOTALL)
+    if not args_match:
+        return name, "", True
+
+    raw = args_match.group(1).rstrip()
+    if complete and raw.endswith("}"):
+        raw = raw[:-1].rstrip()
+    if raw.startswith("{"):
+        inner = raw[1:].rstrip()
+        if inner.endswith("}"):
+            inner = inner[:-1].rstrip()
+        raw = inner
+    return name, raw, True
+
+
+def _find_tool_calls(text: str, start_pos: int = 0):
+    """Find all complete ``{...}`` tool-call objects in *text*.
+
+    Returns a list of dicts with keys *start*, *end*, *name*, *args*,
+    *complete*.
+    """
+    results = []
+    pos = start_pos
+
+    while True:
+        brace = text.find("{", pos)
+        if brace == -1:
+            break
+
+        end, complete = _scan_json(text, brace)
+        if not complete:
+            break
+
+        json_str = text[brace:end]
+        if not _TOOL_CALL_HEAD_RE.search(json_str):
+            pos = end
+            continue
+
+        name, args, valid = _parse_tool_call_json(json_str, complete=True)
+        if not valid or name is None:
+            pos = end
+            continue
+
+        results.append(
+            {
+                "start": brace,
+                "end": end,
+                "name": name,
+                "args": args,
+                "complete": True,
+            }
+        )
+        pos = end
+
+    return results
+
+
+def _find_partial_tool_call(text: str, start_pos: int = 0):
+    """Find one incomplete (still-generating) tool-call JSON object."""
+    brace = text.find("{", start_pos)
+    if brace == -1:
+        return None
+
+    json_str = text[brace:]
+    if not _TOOL_CALL_HEAD_RE.search(json_str):
+        return None
+
+    name, args, valid = _parse_tool_call_json(json_str, complete=False)
+    if not valid or name is None:
+        return None
+
+    return {
+        "start": brace,
+        "name": name,
+        "args": args,
+        "complete": False,
+    }
+
+
+@ToolParserFactory.register("simple_json")
+class SimpleJsonToolParser(BaseToolParser):
+    """Parser for models that output tool calls as plain JSON objects.
+
+    Detects ``{"name": "<func>", "arguments": {...}}`` anywhere in the
+    generated text.  Handles single and (non-overlapping) multiple tool
+    calls.  Text preceding the first tool call is emitted as plain
+    ``content`` deltas.
+    """
+
+    def __init__(self, tools=None, tool_choice="auto"):
+        super().__init__(tools, tool_choice)
+        self._emitted_content_len = 0
+        self._tc_state: List[Dict] = []
+        self._has_tool_calls = False
+
+    # -------------------------------------------------------------- feed
+
+    def feed(
+        self,
+        body: str,
+        current_token_ids: Optional[List[int]] = None,
+        delta_token_ids: Optional[List[int]] = None,
+    ) -> List[Dict]:
+        deltas: List[Dict] = []
+
+        completed = _find_tool_calls(body)
+
+        if not completed:
+            partial = _find_partial_tool_call(body)
+            if not partial:
+                return self._emit_plain_content(body, deltas)
+            all_tcs = [partial]
+        else:
+            all_tcs = completed
+            partial = _find_partial_tool_call(body, completed[-1]["end"])
+            if partial:
+                all_tcs = completed + [partial]
+
+        first_start = all_tcs[0]["start"]
+        if first_start > self._emitted_content_len:
+            content = body[self._emitted_content_len : first_start]
+            self._emitted_content_len = first_start
+            if content:
+                deltas.append({"content": content})
+
+        for i, tc in enumerate(all_tcs):
+            if i >= len(self._tc_state):
+                self._tc_state.append(
+                    {
+                        "id": f"call_{uuid.uuid4().hex[:12]}",
+                        "name_emitted": False,
+                        "args_emitted_len": 0,
+                    }
+                )
+                self._has_tool_calls = True
+            st = self._tc_state[i]
+
+            if not st["name_emitted"]:
+                st["name_emitted"] = True
+                deltas.append(
+                    {
+                        "tool_calls": [
+                            {
+                                "index": i,
+                                "id": st["id"],
+                                "type": "function",
+                                "function": {"name": tc["name"], "arguments": ""},
+                            }
+                        ]
+                    }
+                )
+
+            new_args = tc["args"]
+            if len(new_args) > st["args_emitted_len"]:
+                diff = new_args[st["args_emitted_len"] :]
+                st["args_emitted_len"] = len(new_args)
+                deltas.append(
+                    {
+                        "tool_calls": [
+                            {
+                                "index": i,
+                                "function": {"arguments": diff},
+                            }
+                        ]
+                    }
+                )
+
+        return deltas
+
+    def _emit_plain_content(self, body: str, deltas: List[Dict]) -> List[Dict]:
+        new_content = body[self._emitted_content_len :]
+        if new_content:
+            self._emitted_content_len = len(body)
+            deltas.append({"content": new_content})
+        return deltas
+
+    # -------------------------------------------------------- complete
+
+    def parse_complete(self, body: str) -> Optional[Dict]:
+        completed = _find_tool_calls(body)
+        if not completed:
+            return None
+
+        content = body[: completed[0]["start"]].strip() or None
+        tool_calls = []
+        for i, tc in enumerate(completed):
+            tool_calls.append(
+                {
+                    "id": f"call_{uuid.uuid4().hex[:12]}",
+                    "type": "function",
+                    "function": {
+                        "name": tc["name"],
+                        "arguments": tc["args"],
+                    },
+                }
+            )
+        return {"content": content, "tool_calls": tool_calls}
+
+    @property
+    def has_tool_calls(self) -> bool:
+        return self._has_tool_calls

astrai/inference/core/__init__.py

@@ -2,8 +2,12 @@
 
 from astrai.inference.core.cache import (
     Allocator,
+    CacheView,
+    ContiguousCache,
+    ContiguousCacheView,
     KVCache,
-    KvcacheView,
+    PageCache,
+    PageCacheView,
     PagePool,
     PrefixCache,
     Storage,
@@ -16,8 +20,12 @@ from astrai.inference.core.task import STOP, Task, TaskManager, TaskStatus
 
 __all__ = [
     "Allocator",
+    "CacheView",
     "KVCache",
-    "KvcacheView",
+    "ContiguousCache",
+    "ContiguousCacheView",
+    "PageCache",
+    "PageCacheView",
     "PagePool",
     "PrefixCache",
     "Storage",

astrai/inference/core/cache.py

@@ -1,4 +1,5 @@
 import threading
+from abc import ABC, abstractmethod
 from collections import OrderedDict
 from typing import Callable, Dict, List, Optional, Tuple
 
@@ -62,7 +63,8 @@ class Allocator:
 
     def touch(self, idx: int):
         with self._lock:
-            self._lru.move_to_end(idx)
+            if idx in self._lru:
+                self._lru.move_to_end(idx)
 
 
 class PrefixCache:
@@ -274,7 +276,42 @@ class Storage:
         return k, v
 
 
-class KvcacheView:
+class CacheView(ABC):
+    """Abstract view passed to attention layers for KV-cache I/O."""
+
+    @abstractmethod
+    def write(self, layer_id: int, k: Tensor, v: Tensor): ...
+
+    @abstractmethod
+    def gather(self, layer_id: int) -> Tuple[Tensor, Tensor]: ...
+
+
+class KVCache(ABC):
+    """Abstract KV-cache facade for scheduler/executor."""
+
+    @abstractmethod
+    def task_alloc(self, task_id: str, prompt_ids: List[int]) -> bool: ...
+
+    @abstractmethod
+    def task_free(self, task_id: str): ...
+
+    @abstractmethod
+    def task_extend(self, task_id: str, pos: int) -> bool: ...
+
+    @abstractmethod
+    def bind_tasks(
+        self, task_ids: List[str], total_len: int, device: torch.device
+    ) -> CacheView: ...
+
+    def task_cached(self, task_id: str) -> int:
+        return 0
+
+    def task_record_hashes(
+        self, task_id: str, prompt_ids: List[int], start_logical_page: int = 0
+    ): ...
+
+
+class PageCacheView(CacheView):
     """Bundles Storage + page_table + total_len for attention layers."""
 
     def __init__(self, storage: Storage, page_table: Tensor, total_len: int = 0):
@@ -290,8 +327,8 @@ class KvcacheView:
         return self._storage.gather(layer_id, self._page_table, self._total_len)
 
 
-class KVCache:
-    """Facade: page management + KV-cache I/O for continuous batching."""
+class PageCache(KVCache):
+    """Paged KV-cache with prefix sharing."""
 
     def __init__(
         self,
@@ -361,8 +398,102 @@ class KVCache:
         for i in range(start_logical_page, full_pages):
             self._pool.record(page_table[i], prompt_ids, i)
 
-    def make_table_tensor(self, task_ids: List[str], device: torch.device) -> Tensor:
-        return self._table.table_tensor(task_ids, device)
+    def bind_tasks(
+        self, task_ids: List[str], total_len: int, device: torch.device
+    ) -> PageCacheView:
+        page_table = self._table.table_tensor(task_ids, device)
+        return PageCacheView(self._storage, page_table, total_len)
 
-    def bind(self, page_table: Tensor, total_len: int = 0) -> KvcacheView:
-        return KvcacheView(self._storage, page_table, total_len)
+
+class ContiguousCacheView(CacheView):
+    """Contiguous KV-cache view for attention layers."""
+
+    def __init__(
+        self, cache: "ContiguousCache", batch_indices: Tensor, total_len: int = 0
+    ):
+        self._cache = cache
+        self._batch_indices = batch_indices
+        self._total_len = total_len
+
+    def write(self, layer_id: int, k: Tensor, v: Tensor):
+        seq_len = k.size(1)
+        start_pos = self._total_len - seq_len
+        indices = self._batch_indices
+        self._cache.k[layer_id, indices, start_pos : start_pos + seq_len] = k
+        self._cache.v[layer_id, indices, start_pos : start_pos + seq_len] = v
+        new_len = start_pos + seq_len
+        for s in indices.tolist():
+            cur = self._cache._slot_len.get(s, 0)
+            if new_len > cur:
+                self._cache._slot_len[s] = new_len
+
+    def gather(self, layer_id: int) -> Tuple[Tensor, Tensor]:
+        max_len = max(
+            self._cache._slot_len.get(int(s), 0) for s in self._batch_indices.tolist()
+        )
+        indices = self._batch_indices
+        k = self._cache.k[layer_id, indices, :max_len]
+        v = self._cache.v[layer_id, indices, :max_len]
+        return k, v
+
+
+class ContiguousCache(KVCache):
+    """Contiguous per-slot KV cache (default implementation)."""
+
+    def __init__(
+        self,
+        n_layers: int,
+        max_batch_size: int,
+        max_seq_len: int,
+        n_kv_heads: int,
+        head_dim: int,
+        device: torch.device,
+        dtype: torch.dtype,
+    ):
+        self.max_seq_len = max_seq_len
+        self.k = torch.zeros(
+            n_layers,
+            max_batch_size,
+            max_seq_len,
+            n_kv_heads,
+            head_dim,
+            device=device,
+            dtype=dtype,
+        )
+        self.v = torch.zeros(
+            n_layers,
+            max_batch_size,
+            max_seq_len,
+            n_kv_heads,
+            head_dim,
+            device=device,
+            dtype=dtype,
+        )
+        self._slot_len: Dict[int, int] = {}
+        self._task_slot: Dict[str, int] = {}
+        self._free_slots = list(range(max_batch_size))
+        self._device = device
+
+    def task_alloc(self, task_id: str, prompt_ids: List[int]) -> bool:
+        if not self._free_slots:
+            return False
+        slot = self._free_slots.pop(0)
+        self._task_slot[task_id] = slot
+        self._slot_len[slot] = 0
+        return True
+
+    def task_free(self, task_id: str):
+        slot = self._task_slot.pop(task_id, None)
+        if slot is not None:
+            self._slot_len.pop(slot, None)
+            self._free_slots.append(slot)
+
+    def task_extend(self, task_id: str, pos: int) -> bool:
+        return pos < self.max_seq_len
+
+    def bind_tasks(
+        self, task_ids: List[str], total_len: int, device: torch.device
+    ) -> ContiguousCacheView:
+        slots = [self._task_slot[tid] for tid in task_ids]
+        batch_indices = torch.tensor(slots, dtype=torch.long, device=device)
+        return ContiguousCacheView(self, batch_indices, total_len)

astrai/inference/core/executor.py

@@ -19,13 +19,13 @@ class Executor:
         self,
         model: AutoModel,
         tokenizer: AutoTokenizer,
-        page_cache: KVCache,
+        kv_cache: KVCache,
         device: Optional[str] = None,
         dtype: Optional[torch.dtype] = None,
     ):
         self.model = model
         self.tokenizer = tokenizer
-        self.page_cache = page_cache
+        self.kv_cache = kv_cache
         self.device = device or next(model.parameters()).device
         self.dtype = dtype or next(model.parameters()).dtype
 
@@ -43,7 +43,6 @@ class Executor:
         )
 
         task_ids = [t.task_id for t in tasks]
-        page_tables = self.page_cache.make_table_tensor(task_ids, self.device)
 
         with torch.inference_mode():
             self.model(
@@ -53,7 +52,7 @@ class Executor:
                 )
                 .unsqueeze(0)
                 .expand(batch_sz, -1),
-                paged_cache=self.page_cache.bind(page_tables, total_len=prompt_len),
+                paged_cache=self.kv_cache.bind_tasks(task_ids, prompt_len, self.device),
             )
 
     def execute_decode(self, tasks: List[Task]) -> List[int]:
@@ -72,7 +71,6 @@ class Executor:
         total_len = position_ids.max().item() + 1
 
         task_ids = [t.task_id for t in tasks]
-        page_tables = self.page_cache.make_table_tensor(task_ids, self.device)
 
         temperatures = torch.tensor([t.temperature for t in tasks], device=self.device)
         top_ks = torch.tensor([t.top_k for t in tasks], device=self.device)
@@ -81,7 +79,7 @@ class Executor:
         with torch.inference_mode():
             outputs = self.model(
                 input_ids.unsqueeze(1),
-                paged_cache=self.page_cache.bind(page_tables, total_len=total_len),
+                paged_cache=self.kv_cache.bind_tasks(task_ids, total_len, self.device),
                 position_ids=position_ids.unsqueeze(1),
             )
             logits = outputs["logits"][:, -1, :]

astrai/inference/core/scheduler.py

@@ -4,7 +4,7 @@ from typing import Any, Dict, List, Optional, Tuple
 
 import torch
 
-from astrai.inference.core.cache import KVCache
+from astrai.inference.core.cache import ContiguousCache, KVCache
 from astrai.inference.core.executor import Executor
 from astrai.inference.core.task import STOP, Task, TaskManager, TaskStatus
 from astrai.model.automodel import AutoModel
@@ -14,7 +14,7 @@ logger = logging.getLogger(__name__)
 
 
 class InferenceScheduler:
-    """Four-phase continuous batching loop: cleanup -> refill -> prefill -> decode."""
+    """Continuous batching loop: cleanup -> refill -> prefill -> decode (all groups)."""
 
     def __init__(
         self,
@@ -23,9 +23,9 @@ class InferenceScheduler:
         max_batch_size: int = 16,
         max_seq_len: Optional[int] = None,
         max_prompt_len: int = 2048,
-        page_size: int = 64,
         device: Optional[str] = None,
         dtype: Optional[torch.dtype] = None,
+        cache: Optional[KVCache] = None,
     ):
         config = model.config
 
@@ -41,19 +41,20 @@ class InferenceScheduler:
         self.device = device or next(model.parameters()).device
         self.dtype = dtype or next(model.parameters()).dtype
 
-        n_pages = (
-            max_batch_size * (self.max_seq_len + page_size) + page_size - 1
-        ) // page_size
+        head_dim = config.dim // config.n_heads
 
-        self._page_cache = KVCache(
-            config.n_layers,
-            n_pages,
-            page_size,
-            config.n_kv_heads,
-            config.dim // config.n_heads,
-            self.device,
-            self.dtype,
-        )
+        if cache is not None:
+            self._cache = cache
+        else:
+            self._cache = ContiguousCache(
+                config.n_layers,
+                max_batch_size,
+                self.max_seq_len,
+                config.n_kv_heads,
+                head_dim,
+                self.device,
+                self.dtype,
+            )
 
         self._task_mgr = TaskManager(
             tokenizer=tokenizer,
@@ -65,30 +66,32 @@ class InferenceScheduler:
         self._executor = Executor(
             model=model,
             tokenizer=tokenizer,
-            page_cache=self._page_cache,
+            kv_cache=self._cache,
             device=self.device,
             dtype=self.dtype,
         )
 
-        self._running = False
+        self._stop_event = threading.Event()
+        self._loop_thread: Optional[threading.Thread] = None
 
     def add_task(self, prompt: str, **kwargs) -> str:
         return self._task_mgr.add_task(prompt, **kwargs)
 
     def remove_task(self, task_id: str):
         for task in self._task_mgr.remove_task(task_id):
-            self._page_cache.task_free(task.task_id)
+            self._cache.task_free(task.task_id)
 
     def get_stats(self) -> Dict[str, Any]:
         return self._task_mgr.get_stats()
 
     def _run_generation_loop(self):
         stop_ids = self._task_mgr.tokenizer.stop_ids
+        cache = self._cache
         try:
-            while self._running:
+            while not self._stop_event.is_set():
                 finished = self._task_mgr.remove_finished_tasks(stop_ids)
                 for task in finished:
-                    self._page_cache.task_free(task.task_id)
+                    cache.task_free(task.task_id)
 
                 active = self._task_mgr.get_active_tasks()
                 available = self._task_mgr.max_batch_size - len(active)
@@ -96,7 +99,7 @@ class InferenceScheduler:
                     candidates = self._task_mgr.pull_candidates(available)
                     failed = []
                     for task in candidates:
-                        if self._page_cache.task_alloc(task.task_id, task.prompt_ids):
+                        if cache.task_alloc(task.task_id, task.prompt_ids):
                             self._task_mgr.activate(task)
                         else:
                             failed.append(task)
@@ -111,7 +114,7 @@ class InferenceScheduler:
                     t
                     for t in self._task_mgr.get_active_tasks()
                     if t.output_tokens == 0
-                    and self._page_cache.task_cached(t.task_id) < len(t.prompt_ids)
+                    and cache.task_cached(t.task_id) < len(t.prompt_ids)
                 ]
                 if to_prefill:
                     for t in to_prefill:
@@ -121,36 +124,34 @@ class InferenceScheduler:
                     for t in to_prefill:
                         key = (
                             len(t.prompt_ids),
-                            self._page_cache.task_cached(t.task_id),
+                            cache.task_cached(t.task_id),
                         )
                         groups.setdefault(key, []).append(t)
 
                     for (prompt_len, start_pos), group in groups.items():
                         self._executor.execute_prefill(group, prompt_len, start_pos)
-                        start_logical_page = start_pos // self._page_cache.page_size
+                        start_logical_page = start_pos // getattr(
+                            cache, "page_size", 64
+                        )
                         for t in group:
-                            self._page_cache.task_record_hashes(
-                                t.task_id,
-                                t.prompt_ids,
-                                start_logical_page=start_logical_page,
+                            cache.task_record_hashes(
+                                t.task_id, t.prompt_ids, start_logical_page
                             )
 
                 pos_groups: Dict[int, List[Task]] = {}
                 for t in self._task_mgr.get_active_tasks():
                     pos_groups.setdefault(t.next_pos, []).append(t)
 
-                if pos_groups:
-                    best_key = max(pos_groups, key=lambda k: len(pos_groups[k]))
-                    group = sorted(pos_groups[best_key], key=lambda t: t.task_id)
+                for next_pos in sorted(pos_groups.keys()):
+                    group = sorted(pos_groups[next_pos], key=lambda t: t.task_id)
 
                     valid: List[Task] = []
                     for t in group:
-                        if self._page_cache.task_extend(t.task_id, t.next_pos):
+                        if cache.task_extend(t.task_id, t.next_pos):
                             valid.append(t)
                         else:
                             t.status = TaskStatus.ABORTED
-                            if t.stream_callback:
-                                t.stream_callback(STOP)
+                            self._task_mgr.invoke_callback(t.task_id, STOP)
 
                     if valid:
                         next_tokens = self._executor.execute_decode(valid)
@@ -158,48 +159,44 @@ class InferenceScheduler:
                         for t, ntok in zip(valid, next_tokens):
                             t.output_ids.append(ntok)
                             t.output_tokens += 1
-                            pos = t.input_tokens + t.output_tokens
-                            extend_ok = self._page_cache.task_extend(t.task_id, pos)
-                            if t.stream_callback:
-                                t.stream_callback(
-                                    self._task_mgr.tokenizer.decode([ntok])
-                                )
-                            if not extend_ok:
-                                t.status = TaskStatus.ABORTED
-                                if t.stream_callback:
-                                    t.stream_callback(STOP)
+                            self._task_mgr.invoke_callback(
+                                t.task_id,
+                                self._task_mgr.tokenizer.decode([ntok]),
+                            )
 
                         for t in valid:
                             if t.is_finished(stop_ids):
-                                if t.stream_callback:
-                                    t.stream_callback(STOP)
+                                self._task_mgr.invoke_callback(t.task_id, STOP)
 
         except Exception as e:
+            self._stop_event.set()
             logger.error(f"Scheduler loop crashed: {e}", exc_info=True)
             for task in self._task_mgr.get_active_tasks():
-                if task.stream_callback:
-                    task.stream_callback(STOP)
-                self._page_cache.task_free(task.task_id)
+                self._task_mgr.invoke_callback(task.task_id, STOP)
+                cache.task_free(task.task_id)
             for task in self._task_mgr.get_waiting_tasks():
-                if task.stream_callback:
-                    task.stream_callback(STOP)
+                self._task_mgr.invoke_callback(task.task_id, STOP)
             self._task_mgr.clear_queues()
-            raise
 
     def start(self):
-        if not self._running:
-            self._running = True
-            t = threading.Thread(target=self._run_generation_loop, daemon=True)
-            t.start()
-            self._loop_thread = t
+        if self._loop_thread is not None and self._loop_thread.is_alive():
+            return
+        self._stop_event.clear()
+        t = threading.Thread(target=self._run_generation_loop, daemon=True)
+        t.start()
+        self._loop_thread = t
 
     def stop(self):
-        self._running = False
+        self._stop_event.set()
         self._task_mgr.wake()
-        if hasattr(self, "_loop_thread"):
+        if self._loop_thread is not None:
             self._loop_thread.join(timeout=2.0)
+        self._loop_thread = None
         for task in self._task_mgr.get_active_tasks():
-            self._page_cache.task_free(task.task_id)
+            self._task_mgr.invoke_callback(task.task_id, STOP)
+            self._cache.task_free(task.task_id)
+        for task in self._task_mgr.get_waiting_tasks():
+            self._task_mgr.invoke_callback(task.task_id, STOP)
         self._task_mgr.clear_queues()
         if torch.cuda.is_available():
             torch.cuda.empty_cache()

astrai/inference/core/task.py

@@ -33,7 +33,6 @@ class Task:
         temperature: float = 1.0,
         top_p: float = 1.0,
         top_k: int = 50,
-        stream_callback: Optional[Callable[[str], None]] = None,
     ):
         self.task_id = task_id
         self.prompt_ids = prompt_ids
@@ -48,7 +47,6 @@ class Task:
         self.output_tokens: int = 0
         self.arrival_time = time.time()
         self.finish_time: Optional[float] = None
-        self.stream_callback = stream_callback
 
     @property
     def next_pos(self) -> int:
@@ -79,6 +77,7 @@ class TaskManager:
 
         self.waiting_queue: Deque[Task] = deque()
         self.active_tasks: List[Task] = []
+        self._callbacks: Dict[str, Callable[[str], None]] = {}
 
         self._task_event = threading.Event()
         self._lock = threading.Lock()
@@ -117,12 +116,13 @@ class TaskManager:
             temperature=temperature,
             top_p=top_p,
             top_k=top_k,
-            stream_callback=stream_callback,
         )
 
         with self._lock:
             self.waiting_queue.append(task)
             self._total_tasks += 1
+            if stream_callback:
+                self._callbacks[task_id] = stream_callback
 
         self._task_event.set()
         return task_id
@@ -134,8 +134,14 @@ class TaskManager:
                 t for t in self.waiting_queue if t.task_id != task_id
             )
             self.active_tasks = [t for t in self.active_tasks if t.task_id != task_id]
+            self._callbacks.pop(task_id, None)
         return removed_active
 
+    def invoke_callback(self, task_id: str, token: str):
+        cb = self._callbacks.get(task_id)
+        if cb:
+            cb(token)
+
     def get_stats(self) -> Dict[str, Any]:
         return {
             "total_tasks": self._total_tasks,
@@ -186,7 +192,10 @@ class TaskManager:
         return bool(self.active_tasks or self.waiting_queue)
 
     def wait_for_tasks(self, timeout: float = 1.0):
-        self._task_event.clear()
+        with self._lock:
+            if self.waiting_queue or self.active_tasks:
+                return
+            self._task_event.clear()
         self._task_event.wait(timeout=timeout)
 
     def get_active_tasks(self) -> List[Task]:
@@ -201,6 +210,7 @@ class TaskManager:
         with self._lock:
             self.waiting_queue.clear()
             self.active_tasks.clear()
+            self._callbacks.clear()
 
     def wake(self):
         self._task_event.set()

astrai/inference/engine.py

@@ -8,6 +8,7 @@ from typing import Any, AsyncGenerator, Dict, Generator, List, Optional, Tuple,
 import torch
 import torch.nn as nn
 
+from astrai.inference.core.cache import KVCache
 from astrai.inference.core.scheduler import InferenceScheduler
 from astrai.inference.core.task import STOP
 from astrai.tokenize import AutoTokenizer
@@ -79,8 +80,8 @@ class GenerationRequest:
             raise ValueError("top_k must be a non-negative integer")
         if not (0.0 <= top_p <= 1.0):
             raise ValueError("top_p must be a float between 0.0 and 1.0")
-        if not (isinstance(temperature, (int, float)) and temperature >= 0):
-            raise ValueError("temperature must be a non-negative number")
+        if not (isinstance(temperature, (int, float)) and temperature > 0):
+            raise ValueError("temperature must be a positive number")
 
         self.messages = messages
         self.top_k = top_k
@@ -101,6 +102,7 @@ class InferenceEngine:
         max_seq_len: Optional[int] = None,
         max_prompt_len: int = 2048,
         page_size: int = 128,
+        cache: Optional[KVCache] = None,
     ):
         self.model = model
         self.tokenizer = tokenizer
@@ -110,7 +112,7 @@ class InferenceEngine:
             max_batch_size=max_batch_size,
             max_seq_len=max_seq_len,
             max_prompt_len=max_prompt_len,
-            page_size=page_size,
+            cache=cache,
         )
 
         self.scheduler.start()

astrai/inference/sample.py

@@ -29,6 +29,7 @@ class BaseSamplingStrategy(ABC):
         Returns:
             Transformed logits tensor.
         """
+        raise NotImplementedError
 
 
 class TemperatureStrategy(BaseSamplingStrategy):
@@ -41,13 +42,15 @@ class TemperatureStrategy(BaseSamplingStrategy):
     def __init__(self, temperature: Union[float, Tensor] = 1.0):
         self.temperature = temperature
 
-    def apply(self, logits, filter_value=-float("inf")):
+    def apply(self, logits: Tensor, filter_value: float = -float("inf")) -> Tensor:
         t = self.temperature
         if isinstance(t, Tensor):
+            t = t.to(logits.device, non_blocking=True).view(-1, 1)
+            t = torch.clamp(t, min=1e-8)
             if (t != 1.0).any():
-                logits = logits / t.to(logits.device, non_blocking=True).view(-1, 1)
+                logits = logits / t
         elif t != 1.0:
-            logits = logits / t
+            logits = logits / max(t, 1e-8)
         return logits
 
 
@@ -61,7 +64,7 @@ class TopKStrategy(BaseSamplingStrategy):
     def __init__(self, top_k: Union[int, Tensor] = 0):
         self.top_k = top_k
 
-    def apply(self, logits, filter_value=-float("inf")):
+    def apply(self, logits: Tensor, filter_value: float = -float("inf")) -> Tensor:
         tk = self.top_k
         if isinstance(tk, Tensor):
             tk = tk.to(logits.device, non_blocking=True).long().clamp(min=0)
@@ -98,7 +101,9 @@ class TopPStrategy(BaseSamplingStrategy):
     def __init__(self, top_p: Union[float, Tensor] = 1.0):
         self.top_p = top_p
 
-    def _apply(self, logits, top_p, filter_value):
+    def _apply(
+        self, logits: Tensor, top_p: Union[float, Tensor], filter_value: float
+    ) -> Tensor:
         sorted_logits, sorted_indices = torch.sort(logits, descending=True, dim=-1)
         cum_probs = torch.cumsum(torch.softmax(sorted_logits, dim=-1), dim=-1)
         remove = cum_probs > top_p
@@ -109,7 +114,7 @@ class TopPStrategy(BaseSamplingStrategy):
         logits[mask] = filter_value
         return logits
 
-    def apply(self, logits, filter_value=-float("inf")):
+    def apply(self, logits: Tensor, filter_value: float = -float("inf")) -> Tensor:
         tp = self.top_p
         if isinstance(tp, Tensor):
             tp = tp.to(logits.device, non_blocking=True)
@@ -140,7 +145,7 @@ class SamplingPipeline(BaseSamplingStrategy):
     def __init__(self, strategies: List[BaseSamplingStrategy]):
         self.strategies = strategies
 
-    def apply(self, logits, filter_value=-float("inf")):
+    def apply(self, logits: Tensor, filter_value: float = -float("inf")) -> Tensor:
         for strategy in self.strategies:
             logits = strategy.apply(logits, filter_value)
         return logits

astrai/model/components/attention.py

@@ -6,7 +6,7 @@ import torch.nn.functional as F
 from torch import Tensor
 
 from astrai.factory import BaseFactory
-from astrai.inference.core.cache import KvcacheView
+from astrai.inference.core.cache import CacheView
 from astrai.model.components.linear import Linear
 from astrai.model.components.norm import RMSNorm
 from astrai.model.components.rope import apply_rotary_emb
@@ -24,9 +24,7 @@ def repeat_kv(x: Tensor, n_rep: int) -> Tensor:
 
 
 class AttnFactory(BaseFactory[nn.Module]):
-    @classmethod
-    def create(cls, attn_type: str, **kwargs) -> nn.Module:
-        return super().create(attn_type, **kwargs)
+    pass
 
 
 @AttnFactory.register("gqa")
@@ -40,6 +38,7 @@ class GQA(nn.Module):
         norm_eps: float,
         use_gated_attention: bool,
         layer_id: int,
+        n_layers: int = 1,
     ):
         super().__init__()
         assert dim % n_heads == 0
@@ -57,7 +56,7 @@ class GQA(nn.Module):
         self.q_proj = Linear(dim, n_heads * self.head_dim)
         self.k_proj = Linear(dim, n_kv_heads * self.head_dim)
         self.v_proj = Linear(dim, n_kv_heads * self.head_dim)
-        self.o_proj = Linear(dim, dim)
+        self.o_proj = Linear(dim, dim, init_std=0.02 / (2 * n_layers) ** 0.5)
 
         if self.use_qk_norm:
             self.q_norm = RMSNorm(self.head_dim, norm_eps)
@@ -76,7 +75,7 @@ class GQA(nn.Module):
         x: Tensor,
         rotary_emb: Tensor,
         attn_mask: Tensor = None,
-        paged_cache: Optional[KvcacheView] = None,
+        paged_cache: Optional[CacheView] = None,
     ) -> Tensor:
         is_causal = attn_mask is None
 
@@ -123,6 +122,7 @@ class MLA(nn.Module):
         use_qk_norm: bool,
         use_gated_attention: bool,
         layer_id: int,
+        n_layers: int = 1,
     ):
         super().__init__()
         self.dim = dim
@@ -150,7 +150,9 @@ class MLA(nn.Module):
             n_kv_heads * (2 * self.head_dim),
         )
 
-        self.o_proj = Linear(dim, dim, bias=False)
+        self.o_proj = Linear(
+            dim, dim, bias=False, init_std=0.02 / (2 * n_layers) ** 0.5
+        )
 
         if use_gated_attention:
             self.gate = Linear(dim, dim, bias=False)
@@ -160,7 +162,7 @@ class MLA(nn.Module):
         x: Tensor,
         rotary_emb: Tensor,
         attn_mask: Tensor = None,
-        paged_cache: Optional[KvcacheView] = None,
+        paged_cache: Optional[CacheView] = None,
     ) -> Tensor:
         bsz, seq_len, _ = x.size()
         is_causal = attn_mask is None

astrai/model/components/decoder_block.py

@@ -1,51 +1,31 @@
+from dataclasses import asdict
 from typing import Optional
 
 import torch.nn as nn
 from torch import Tensor
 
-from astrai.inference.core.cache import KvcacheView
+from astrai.inference.core.cache import CacheView
 from astrai.model.components.attention import AttnFactory
 from astrai.model.components.mlp import FFNFactory
 from astrai.model.components.norm import RMSNorm
 
 
 class DecoderBlock(nn.Module):
-    def __init__(
-        self,
-        dim: int,
-        n_heads: int,
-        dim_ffn: int,
-        n_kv_heads: int,
-        norm_eps: float,
-        use_qk_norm: bool,
-        use_gated_attention: bool,
-        layer_id: int,
-        attn_type: str = "gqa",
-        ffn_type: str = "mlp",
-        **kwargs,
-    ):
+    def __init__(self, config, layer_id: int):
         super().__init__()
-        self.attention = AttnFactory.create(
-            attn_type,
-            dim=dim,
-            n_heads=n_heads,
-            n_kv_heads=n_kv_heads,
-            use_qk_norm=use_qk_norm,
-            norm_eps=norm_eps,
-            use_gated_attention=use_gated_attention,
-            layer_id=layer_id,
-            **kwargs,
-        )
-        self.input_norm = RMSNorm(dim, norm_eps)
-        self.post_attention_norm = RMSNorm(dim, norm_eps)
-        self.mlp = FFNFactory.create(ffn_type, dim, dim_ffn, **kwargs)
+        cfg = asdict(config)
+        cfg["down_init_std"] = 0.02 / (2 * config.n_layers) ** 0.5
+        self.attention = AttnFactory.create(config.attn_type, **cfg, layer_id=layer_id)
+        self.input_norm = RMSNorm(config.dim, config.norm_eps)
+        self.post_attention_norm = RMSNorm(config.dim, config.norm_eps)
+        self.mlp = FFNFactory.create(config.ffn_type, **cfg)
 
     def forward(
         self,
         x: Tensor,
         rotary_emb: Tensor,
         attention_mask: Optional[Tensor] = None,
-        paged_cache: Optional[KvcacheView] = None,
+        paged_cache: Optional[CacheView] = None,
     ) -> Tensor:
         attn_output = self.attention(
             self.input_norm(x),

astrai/model/components/embedding.py

@@ -1,3 +1,5 @@
+import math
+
 import torch
 import torch.nn as nn
 import torch.nn.functional as F
@@ -5,12 +7,20 @@ from torch import Tensor
 
 
 class Embedding(nn.Module):
-    def __init__(self, vocab_size: int, embedding_dim: int):
+    def __init__(self, vocab_size: int, embedding_dim: int, neftune_alpha: float = 0.0):
         super().__init__()
         self.weight = nn.Parameter(torch.empty((vocab_size, embedding_dim)))
+        self.neftune_noise_alpha = neftune_alpha
+
+    def set_neftune_alpha(self, alpha: float):
+        self.neftune_noise_alpha = alpha
 
     def reset_parameters(self):
         nn.init.normal_(self.weight, mean=0.0, std=0.02)
 
     def forward(self, x: Tensor) -> Tensor:
-        return F.embedding(x, self.weight)
+        out = F.embedding(x, self.weight)
+        if self.training and self.neftune_noise_alpha > 0.0:
+            eps = self.neftune_noise_alpha / math.sqrt(out.size(1))
+            out = out + eps * torch.randn_like(out)
+        return out

astrai/model/components/linear.py

@@ -5,13 +5,16 @@ from torch import Tensor
 
 
 class Linear(nn.Module):
-    def __init__(self, in_dim: int, out_dim: int, bias: bool = False):
+    def __init__(
+        self, in_dim: int, out_dim: int, bias: bool = False, init_std: float = 0.02
+    ):
         super().__init__()
         self.weight = nn.Parameter(torch.empty((out_dim, in_dim)))
         self.bias = nn.Parameter(torch.zeros(out_dim)) if bias else None
+        self.init_std = init_std
 
     def reset_parameters(self):
-        nn.init.kaiming_uniform_(self.weight, a=5**0.5)
+        nn.init.normal_(self.weight, mean=0.0, std=self.init_std)
         if self.bias is not None:
             fan_in, _ = nn.init._calculate_fan_in_and_fan_out(self.weight)
             bound = 1 / (fan_in**0.5)

astrai/model/components/mlp.py

@@ -8,18 +8,16 @@ from astrai.model.components.linear import Linear
 
 
 class FFNFactory(BaseFactory[nn.Module]):
-    @classmethod
-    def create(cls, ffn_type: str, dim: int, dim_ffn: int, **kwargs) -> nn.Module:
-        return super().create(ffn_type, dim, dim_ffn, **kwargs)
+    pass
 
 
 @FFNFactory.register("mlp")
 class MLP(nn.Module):
-    def __init__(self, dim: int, dim_ffn: int):
+    def __init__(self, dim: int, dim_ffn: int, down_init_std: float = 0.02):
         super().__init__()
         self.up = Linear(dim, dim_ffn)
         self.gate = Linear(dim, dim_ffn)
-        self.down = Linear(dim_ffn, dim)
+        self.down = Linear(dim_ffn, dim, init_std=down_init_std)
 
     def forward(self, x: Tensor) -> Tensor:
         gated = self.up(x) * F.silu(self.gate(x))
@@ -37,6 +35,7 @@ class DeepSeekMoE(nn.Module):
         n_shared_experts: int = 1,
         n_activated_experts: int = 2,
         topk_method: str = "greedy",
+        n_layers: int = 1,
     ):
         super().__init__()
         self.dim = dim
@@ -46,12 +45,20 @@ class DeepSeekMoE(nn.Module):
         self.topk_method = topk_method
 
         self.router = Linear(dim, n_routed_experts, bias=False)
+        moe_scale = 1 / max(n_shared_experts, 1) + 1 / n_activated_experts
+        down_init_std = 0.02 / (2 * n_layers * moe_scale) ** 0.5
 
         self.shared_experts = nn.ModuleList(
-            [MLP(dim, dim_ffn) for _ in range(n_shared_experts)]
+            [
+                MLP(dim, dim_ffn, down_init_std=down_init_std)
+                for _ in range(n_shared_experts)
+            ]
         )
         self.routed_experts = nn.ModuleList(
-            [MLP(dim, dim_ffn) for _ in range(n_routed_experts)]
+            [
+                MLP(dim, dim_ffn, down_init_std=down_init_std)
+                for _ in range(n_routed_experts)
+            ]
         )
 
     def forward(self, x: Tensor) -> Tensor:

astrai/model/encoder.py

@@ -23,22 +23,12 @@ class EmbeddingEncoder(AutoModel):
         self.rotary_embedding = RotaryEmbedding(
             rope_dim, config.max_len, rope_base, rope_scaling=config.rope_scaling
         )
-        self.embed_tokens = Embedding(config.vocab_size, config.dim)
+        self.embed_tokens = Embedding(
+            config.vocab_size, config.dim, neftune_alpha=config.neftune_alpha
+        )
 
         self.layers = nn.ModuleList(
-            [
-                DecoderBlock(
-                    config.dim,
-                    config.n_heads,
-                    config.dim_ffn,
-                    config.n_kv_heads,
-                    config.norm_eps,
-                    config.use_qk_norm,
-                    config.use_gated_attention,
-                    layer_id,
-                )
-                for layer_id in range(config.n_layers)
-            ]
+            [DecoderBlock(config, layer_id) for layer_id in range(config.n_layers)]
         )
 
         self.norm = RMSNorm(config.dim, config.norm_eps)

astrai/model/transformer.py

@@ -5,7 +5,7 @@ import torch.nn as nn
 from torch import Tensor
 
 from astrai.config.model_config import AutoRegressiveLMConfig
-from astrai.inference.core.cache import KvcacheView
+from astrai.inference.core.cache import CacheView
 from astrai.model.automodel import AutoModel
 from astrai.model.components.decoder_block import DecoderBlock
 from astrai.model.components.embedding import Embedding
@@ -26,24 +26,21 @@ def process_attention_mask(
         return input_mask
 
     device = input_tensor.device
-    dtype = input_tensor.dtype
-    B, S = input_tensor.size()[:2]
+    B = input_tensor.size(0)
     T = position_ids.max().item() + 1
 
     if input_mask is None:
         if position_ids.min().item() == 0 and is_causal:
             return None
-        pad = torch.ones(B, T, dtype=torch.bool, device=device)
+        attend = torch.ones(B, 1, T, dtype=torch.bool, device=device)
     else:
-        pad = input_mask[:, :T].to(device=device, dtype=torch.bool)
+        attend = input_mask[:, :T].to(device=device, dtype=torch.bool).unsqueeze(1)
 
-    attend = pad.view(B, 1, T).expand(B, S, T).clone()
     if is_causal:
-        attend &= position_ids.unsqueeze(-1) >= torch.arange(T, device=device)
+        causal = position_ids.unsqueeze(-1) >= torch.arange(T, device=device)
+        attend = attend & causal
 
-    return torch.full(
-        (B, 1, S, T), -torch.finfo(dtype).max / 2, dtype=dtype, device=device
-    ).masked_fill_(attend.unsqueeze(1), 0.0)
+    return attend.unsqueeze(1)
 
 
 @AutoModel.register("autoregressive_lm")
@@ -62,31 +59,12 @@ class AutoRegressiveLM(AutoModel):
         self.rotary_embedding = RotaryEmbedding(
             rope_dim, config.max_len, rope_base, rope_scaling=config.rope_scaling
         )
-        self.embed_tokens = Embedding(config.vocab_size, config.dim)
+        self.embed_tokens = Embedding(
+            config.vocab_size, config.dim, neftune_alpha=config.neftune_alpha
+        )
 
         self.layers = nn.ModuleList(
-            [
-                DecoderBlock(
-                    config.dim,
-                    config.n_heads,
-                    config.dim_ffn,
-                    config.n_kv_heads,
-                    config.norm_eps,
-                    config.use_qk_norm,
-                    config.use_gated_attention,
-                    layer_id,
-                    attn_type=config.attn_type,
-                    ffn_type=config.ffn_type,
-                    n_routed_experts=config.n_routed_experts,
-                    n_shared_experts=config.n_shared_experts,
-                    n_activated_experts=config.n_activated_experts,
-                    topk_method=config.topk_method,
-                    kv_lora_rank=config.kv_lora_rank,
-                    qk_nope_head_dim=config.qk_nope_head_dim,
-                    qk_rope_head_dim=config.qk_rope_head_dim,
-                )
-                for layer_id in range(config.n_layers)
-            ]
+            [DecoderBlock(config, layer_id) for layer_id in range(config.n_layers)]
         )
 
         self.norm = RMSNorm(config.dim, config.norm_eps)
@@ -134,7 +112,7 @@ class AutoRegressiveLM(AutoModel):
         self,
         input_ids: Tensor,
         input_mask: Optional[Tensor] = None,
-        paged_cache: Optional[KvcacheView] = None,
+        paged_cache: Optional[CacheView] = None,
         position_ids: Optional[Tensor] = None,
     ) -> Dict[str, Tensor]:
         assert input_ids.ndim == 2

astrai/parallel/executor.py

@@ -2,11 +2,13 @@
 
 import contextlib
 import logging
+import os
 from contextlib import contextmanager
 from typing import Optional, Tuple
 
 import torch
 import torch.nn as nn
+from torch.distributed.fsdp import FullStateDictConfig, StateDictType
 from torch.distributed.fsdp import FullyShardedDataParallel as FSDP
 from torch.nn.parallel import DistributedDataParallel as DDP
 from torch.optim import Optimizer
@@ -115,8 +117,8 @@ class BaseExecutor:
     def backward(self, loss: torch.Tensor):
         loss.backward()
 
-    def unwrap_model(self, model: nn.Module) -> nn.Module:
-        return model
+    def unwrap_model(self, model: nn.Module):
+        return model.state_dict()
 
     @property
     def use_distributed(self) -> bool:
@@ -130,6 +132,12 @@ class BaseExecutor:
     def grad_accum_steps(self) -> int:
         return self.gradient_state.num_steps
 
+    def clip_grad_norm(self, model: nn.Module, max_norm: float) -> float:
+        total_norm = torch.nn.utils.clip_grad_norm_(model.parameters(), max_norm)
+        if isinstance(total_norm, torch.Tensor):
+            return total_norm.item()
+        return total_norm
+
 
 class ExecutorFactory(BaseFactory[BaseExecutor]):
     pass
@@ -180,7 +188,7 @@ class DDPExecutor(BaseExecutor):
         if not self.use_distributed:
             logger.warning("DDP backend selected but world_size=1, model not wrapped")
             return model
-        local_rank = get_rank()
+        local_rank = int(os.environ.get("LOCAL_RANK", get_rank()))
         model = DDP(
             model,
             device_ids=[local_rank],
@@ -195,10 +203,10 @@ class DDPExecutor(BaseExecutor):
             return model.no_sync()
         return contextlib.nullcontext()
 
-    def unwrap_model(self, model: nn.Module) -> nn.Module:
+    def unwrap_model(self, model: nn.Module):
         if isinstance(model, DDP):
-            return model.module
-        return model
+            return model.module.state_dict()
+        return model.state_dict()
 
 
 @ExecutorFactory.register("fsdp")
@@ -217,7 +225,6 @@ class FSDPExecutor(BaseExecutor):
         sync_module_states: bool = False,
         forward_prefetch: bool = False,
         limit_all_gathers: bool = True,
-        use_orig_params: bool = False,
         ignored_states=None,
         device_mesh=None,
     ):
@@ -236,7 +243,7 @@ class FSDPExecutor(BaseExecutor):
                 sync_module_states=sync_module_states,
                 forward_prefetch=forward_prefetch,
                 limit_all_gathers=limit_all_gathers,
-                use_orig_params=use_orig_params,
+                use_orig_params=True,
                 ignored_states=ignored_states,
                 device_mesh=device_mesh,
             ).items()
@@ -259,9 +266,21 @@ class FSDPExecutor(BaseExecutor):
             return model.no_sync()
         return contextlib.nullcontext()
 
-    def unwrap_model(self, model: nn.Module) -> nn.Module:
-        if self._original_model is not None:
-            return self._original_model
-        if isinstance(model, FSDP):
-            return model._fsdp_wrapped_module
-        return model
+    def clip_grad_norm(self, model: nn.Module, max_norm: float) -> float:
+        if isinstance(model, FSDP) and self.use_distributed:
+            total_norm = model.clip_grad_norm_(max_norm)
+            if isinstance(total_norm, torch.Tensor):
+                return total_norm.item()
+            return total_norm
+        return super().clip_grad_norm(model, max_norm)
+
+    def unwrap_model(self, model: nn.Module):
+        if isinstance(model, FSDP) and self.use_distributed:
+            with FSDP.state_dict_type(
+                model,
+                StateDictType.FULL_STATE_DICT,
+                FullStateDictConfig(offload_to_cpu=True, rank0_only=False),
+            ):
+                return model.state_dict()
+
+        return model.state_dict()

astrai/parallel/setup.py

@@ -1,4 +1,5 @@
 import os
+from abc import ABC, abstractmethod
 from contextlib import contextmanager
 from functools import wraps
 from typing import Callable
@@ -30,6 +31,7 @@ def get_rank() -> int:
 def setup_parallel(
     rank: int,
     world_size: int,
+    local_rank: int,
     backend: str = "nccl",
     master_addr: str = "localhost",
     master_port: str = "29500",
@@ -41,20 +43,26 @@ def setup_parallel(
         return
 
     if world_size <= 1:
+        device_id = torch.device(device_type, local_rank)
+        os.environ["LOCAL_RANK"] = str(local_rank)
+        os.environ["WORLD_SIZE"] = "1"
+        os.environ["LOCAL_DEVICE"] = str(device_id)
         yield None
         return
 
-    device_id = torch.device(device_type, rank)
+    device_id = torch.device(device_type, local_rank)
 
     os.environ["MASTER_ADDR"] = master_addr
     os.environ["MASTER_PORT"] = master_port
-    os.environ["LOCAL_RANK"] = str(rank)
+    os.environ["LOCAL_RANK"] = str(local_rank)
     os.environ["WORLD_SIZE"] = str(world_size)
     os.environ["LOCAL_DEVICE"] = str(device_id)
 
-    dist.init_process_group(
-        rank=rank, world_size=world_size, backend=backend, device_id=device_id
-    )
+    pg_kwargs = dict(rank=rank, world_size=world_size, backend=backend)
+    if backend in ("nccl", "ccl"):
+        pg_kwargs["device_id"] = device_id
+
+    dist.init_process_group(**pg_kwargs)
 
     try:
         if backend == "nccl" and torch.cuda.is_available():
@@ -90,7 +98,7 @@ def only_on_rank(rank, sync=False):
     return decorator
 
 
-def wrapper_spawn_func(
+def _run_single_rank(
     rank: int,
     world_size: int,
     backend: str,
@@ -100,20 +108,108 @@ def wrapper_spawn_func(
     func: Callable,
     kwargs: dict,
 ):
-    try:
+    with setup_parallel(
+        rank=rank,
+        world_size=world_size,
+        local_rank=rank,
+        backend=backend,
+        master_addr=master_addr,
+        master_port=master_port,
+        device_type=device_type,
+    ):
+        func(**kwargs)
+
+
+class LaunchStrategy(ABC):
+    """Strategy for launching a function in a distributed context."""
+
+    def __init__(
+        self,
+        world_size: int,
+        backend: str,
+        master_addr: str,
+        master_port: str,
+        device_type: str,
+        start_method: str,
+    ):
+        self.world_size = world_size
+        self.backend = backend
+        self.master_addr = master_addr
+        self.master_port = master_port
+        self.device_type = device_type
+        self.start_method = start_method
+
+    @abstractmethod
+    def launch(self, func: Callable, **kwargs):
+        raise NotImplementedError
+
+
+class TorchrunStrategy(LaunchStrategy):
+    """External orchestrator (torchrun, SLURM, K8s) — env vars pre-set."""
+
+    def launch(self, func: Callable, **kwargs):
+        rank = int(os.environ["RANK"])
+        world_size = int(os.environ["WORLD_SIZE"])
+        local_rank = int(os.environ.get("LOCAL_RANK", rank))
         with setup_parallel(
             rank=rank,
             world_size=world_size,
-            backend=backend,
-            master_addr=master_addr,
-            master_port=master_port,
-            device_type=device_type,
+            local_rank=local_rank,
+            backend=self.backend,
+            master_addr=os.environ.get("MASTER_ADDR", self.master_addr),
+            master_port=os.environ.get("MASTER_PORT", self.master_port),
+            device_type=self.device_type,
         ):
             func(**kwargs)
 
-    except Exception as e:
-        print(f"Error in rank {rank}: {e}")
-        raise
+
+class LocalStrategy(LaunchStrategy):
+    """Local launcher — single-process or mp.start_processes."""
+
+    def launch(self, func: Callable, **kwargs):
+        args = (
+            self.world_size,
+            self.backend,
+            self.master_addr,
+            self.master_port,
+            self.device_type,
+            func,
+            kwargs,
+        )
+
+        if self.world_size == 1:
+            _run_single_rank(0, *args)
+            return
+
+        ctx = mp.start_processes(
+            _run_single_rank,
+            args=args,
+            nprocs=self.world_size,
+            start_method=self.start_method,
+            join=False,
+        )
+        try:
+            while not ctx.join():
+                pass
+        except BaseException:
+            for p in ctx.processes:
+                p.terminate()
+            ctx.join()
+            raise
+
+
+def _detect_launcher() -> str:
+    """Detect the distributed launcher from environment.
+
+    Returns one of: "torchelastic", "torchrun", "external", "local".
+    """
+    if dist.is_torchelastic_launched():
+        return "torchelastic"
+    if "LOCAL_WORLD_SIZE" in os.environ:
+        return "torchrun"
+    if "RANK" in os.environ and "WORLD_SIZE" in os.environ:
+        return "external"
+    return "local"
 
 
 def spawn_parallel_fn(
@@ -126,41 +222,13 @@ def spawn_parallel_fn(
     start_method: str = "spawn",
     **kwargs,
 ):
-    # clear environment variables
-    for key in [
-        "MASTER_ADDR",
-        "MASTER_PORT",
-        "RANK",
-        "WORLD_SIZE",
-        "LOCAL_RANK",
-        "LOCAL_DEVICE",
-    ]:
-        if key in os.environ:
-            del os.environ[key]
-
-    if world_size == 1:
-        device_id = torch.device(device_type, 0)
-        os.environ["LOCAL_RANK"] = "0"
-        os.environ["WORLD_SIZE"] = "1"
-        os.environ["LOCAL_DEVICE"] = str(device_id)
-
-        func(**kwargs)
-        return
-
-    wrapper_spawn_func_args = (
-        world_size,
-        backend,
-        master_addr,
-        master_port,
-        device_type,
-        func,
-        kwargs,
-    )
-
-    mp.start_processes(
-        wrapper_spawn_func,
-        args=wrapper_spawn_func_args,
-        nprocs=world_size,
-        start_method=start_method,
-        join=True,
-    )
+    launcher = _detect_launcher()
+    if launcher in ("torchelastic", "torchrun", "external"):
+        strategy = TorchrunStrategy(
+            world_size, backend, master_addr, master_port, device_type, start_method
+        )
+    else:
+        strategy = LocalStrategy(
+            world_size, backend, master_addr, master_port, device_type, start_method
+        )
+    strategy.launch(func, **kwargs)

astrai/preprocessing/__init__.py

@@ -0,0 +1,36 @@
+from astrai.preprocessing.builder import (
+    BaseMaskBuilder,
+    MaskBuilderFactory,
+    MultiOutputMaskBuilder,
+    SectionedMaskBuilder,
+    SingleOutputMaskBuilder,
+)
+from astrai.preprocessing.packing import (
+    PackingStrategy,
+    PackingStrategyFactory,
+)
+from astrai.preprocessing.pipeline import Pipeline, filter_by_length
+from astrai.preprocessing.position_id import (
+    PositionIdStrategy,
+    PositionIdStrategyFactory,
+)
+from astrai.preprocessing.writer import (
+    StoreWriter,
+    StoreWriterFactory,
+)
+
+__all__ = [
+    "BaseMaskBuilder",
+    "MaskBuilderFactory",
+    "MultiOutputMaskBuilder",
+    "PackingStrategy",
+    "PackingStrategyFactory",
+    "Pipeline",
+    "PositionIdStrategy",
+    "PositionIdStrategyFactory",
+    "SectionedMaskBuilder",
+    "SingleOutputMaskBuilder",
+    "StoreWriter",
+    "StoreWriterFactory",
+    "filter_by_length",
+]

astrai/preprocessing/builder.py

@@ -0,0 +1,324 @@
+"""Mask building for preprocessing pipeline.
+
+:class:`SectionRenderer` converts section specs into token ids and loss
+masks (template / text / value extraction).  :class:`SingleOutputMaskBuilder`
+handles single-output (SFT / pretrain), :class:`MultiOutputMaskBuilder`
+handles multi-output (DPO / GRPO), and :class:`SectionedMaskBuilder`
+orchestrates both modes as a façade.
+"""
+
+from abc import ABC, abstractmethod
+from typing import Optional
+
+from astrai.factory import BaseFactory
+
+
+def _extract_domain(item: dict, domain_key: Optional[str]) -> str:
+    if not domain_key:
+        return "__default__"
+    val = item.get(domain_key, "__default__")
+    return val if isinstance(val, str) else "__default__"
+
+
+def _resolve_action(action: str, role: str, config) -> str:
+    if action == "$role":
+        return config.mask.get(role, config.mask_default)
+    return action
+
+
+class SectionRenderer:
+    """Render section specs into ``(ids, loss_mask)`` tuples."""
+
+    def process_sections(
+        self,
+        item: dict,
+        sections: list,
+        config,
+        tokenizer,
+        *,
+        is_top_level: bool = False,
+    ):
+        all_ids: list[int] = []
+        loss_mask: list[int] = []
+
+        has_template = any(s.get("template") for s in sections)
+        is_text_config = not has_template and all(
+            s["action"] == "train" for s in sections
+        )
+
+        if is_top_level and has_template and tokenizer.bos_token_id is not None:
+            all_ids.append(tokenizer.bos_token_id)
+            loss_mask.append(0)
+
+        first_section = True
+        for sec in sections:
+            field = sec["field"]
+            action = sec["action"]
+            use_template = sec.get("template", False)
+            add_special = sec.get(
+                "add_special_tokens", not use_template and first_section
+            )
+
+            if use_template:
+                success = self._append_template(
+                    item, field, action, tokenizer, config, all_ids, loss_mask
+                )
+                if not success:
+                    continue
+            else:
+                success = self._append_text(
+                    item,
+                    field,
+                    action,
+                    tokenizer,
+                    add_special,
+                    is_text_config,
+                    config,
+                    all_ids,
+                    loss_mask,
+                )
+                if not success:
+                    continue
+
+            first_section = False
+
+        max_len = config.preprocessing.max_seq_len
+        all_ids = all_ids[:max_len]
+        loss_mask = loss_mask[: len(all_ids)]
+
+        if not all_ids:
+            return None, None
+
+        if is_top_level and has_template and len(all_ids) <= 1:
+            return None, None
+
+        return all_ids, loss_mask
+
+    def process_list_field(self, item: dict, sections: list, config, tokenizer):
+        all_ids: list[int] = []
+        loss_mask: list[int] = []
+
+        for sec in sections:
+            field = sec["field"]
+            action = sec["action"]
+            use_template = sec.get("template", False)
+
+            values = item.get(field)
+            if not isinstance(values, list):
+                continue
+
+            for val in values:
+                if use_template:
+                    if isinstance(val, list):
+                        wrapper = {field: val}
+                        self._append_template(
+                            wrapper,
+                            field,
+                            action,
+                            tokenizer,
+                            config,
+                            all_ids,
+                            loss_mask,
+                        )
+                else:
+                    wrapper = {field: str(val)}
+                    self._append_text(
+                        wrapper,
+                        field,
+                        action,
+                        tokenizer,
+                        False,
+                        False,
+                        config,
+                        all_ids,
+                        loss_mask,
+                    )
+
+        max_len = config.preprocessing.max_seq_len
+        all_ids = all_ids[:max_len]
+        loss_mask = loss_mask[: len(all_ids)]
+
+        if not all_ids:
+            return None, None
+        return all_ids, loss_mask
+
+    @staticmethod
+    def is_value_section(sections: list) -> bool:
+        return len(sections) == 1 and sections[0].get("action") == "value"
+
+    @staticmethod
+    def extract_raw_value(item: dict, sections: list):
+        sec = sections[0]
+        field = sec["field"]
+        raw = item.get(field)
+        if raw is None:
+            return None
+        if isinstance(raw, list):
+            return [float(v) for v in raw]
+        return [float(raw)]
+
+    def _append_template(
+        self, item, field, action, tokenizer, config, all_ids, loss_mask
+    ):
+        messages = item.get(field)
+        if not isinstance(messages, list) or not messages:
+            return False
+        for msg in messages:
+            role = msg.get("role", "")
+            act = _resolve_action(action, role, config)
+            rendered = tokenizer.apply_chat_template(
+                [msg], tokenize=False, add_generation_prompt=False
+            )
+            ids = tokenizer.encode(rendered, add_special_tokens=False)
+            all_ids.extend(ids)
+            val = 1 if act == "train" else 0
+            loss_mask.extend([val] * len(ids))
+        return True
+
+    def _append_text(
+        self,
+        item,
+        field,
+        action,
+        tokenizer,
+        add_special,
+        is_text_config,
+        config,
+        all_ids,
+        loss_mask,
+    ):
+        text = str(item.get(field, ""))
+        if not text.strip():
+            return False
+        if is_text_config:
+            pp = config.preprocessing
+            if pp.min_chars > 0 and len(text) < pp.min_chars:
+                return False
+            if len(text) > pp.max_chars:
+                return False
+        ids = tokenizer.encode(text, add_special_tokens=add_special)
+        all_ids.extend(ids)
+        val = 1 if action == "train" else 0
+        loss_mask.extend([val] * len(ids))
+        return True
+
+
+class BaseMaskBuilder(ABC):
+    """Convert a JSONL item into token ids and optional loss_mask."""
+
+    @abstractmethod
+    def build(self, item: dict, config, tokenizer) -> Optional[dict]: ...
+
+
+class MaskBuilderFactory(BaseFactory["BaseMaskBuilder"]):
+    pass
+
+
+@MaskBuilderFactory.register("single")
+class SingleOutputMaskBuilder(BaseMaskBuilder):
+    """Build a single output sequence with optional loss mask.
+
+    Expects ``config.input.sections`` (list of section specs).
+    """
+
+    def __init__(self, renderer: Optional[SectionRenderer] = None):
+        self.renderer = renderer or SectionRenderer()
+
+    def build(self, item: dict, config, tokenizer) -> Optional[dict]:
+        sections = config.input.sections
+        if not sections:
+            return None
+
+        ids, mask = self.renderer.process_sections(
+            item, sections, config, tokenizer, is_top_level=True
+        )
+        if ids is None:
+            return None
+
+        result: dict = {
+            "sequence": ids,
+            "domain": _extract_domain(item, config.output.domain_key),
+        }
+        if not all(m == 1 for m in mask):
+            result["loss_mask"] = mask
+        return result
+
+
+@MaskBuilderFactory.register("multi")
+class MultiOutputMaskBuilder(BaseMaskBuilder):
+    """Build multiple output sequences (DPO / GRPO).
+
+    Expects ``config.input.sources`` (dict of output_key → spec).
+    """
+
+    def __init__(self, renderer: Optional[SectionRenderer] = None):
+        self.renderer = renderer or SectionRenderer()
+
+    def build(self, item: dict, config, tokenizer) -> Optional[dict]:
+        sources_spec = getattr(config.input, "sources", None)
+        if not sources_spec:
+            return None
+
+        result: dict = {}
+        any_output = False
+
+        for output_key, spec in sources_spec.items():
+            sections = spec.get("sections", [])
+            if not sections:
+                continue
+
+            if self.renderer.is_value_section(sections):
+                ids = self.renderer.extract_raw_value(item, sections)
+                if ids is None:
+                    continue
+                result[output_key] = ids
+                any_output = True
+                continue
+
+            list_field = spec.get("list_field", False)
+            mask_key = spec.get("mask_key", f"{output_key}_mask")
+
+            if list_field:
+                ids, mask = self.renderer.process_list_field(
+                    item, sections, config, tokenizer
+                )
+            else:
+                ids, mask = self.renderer.process_sections(
+                    item, sections, config, tokenizer, is_top_level=True
+                )
+
+            if ids is None:
+                continue
+
+            result[output_key] = ids
+            if not all(m == 1 for m in mask):
+                result[mask_key] = mask
+            elif "mask_key" in spec:
+                result[mask_key] = mask
+
+            any_output = True
+
+        if not any_output:
+            return None
+
+        result["domain"] = _extract_domain(item, config.output.domain_key)
+        return result
+
+
+@MaskBuilderFactory.register("sectioned")
+class SectionedMaskBuilder(BaseMaskBuilder):
+    """Façade that dispatches to SingleOutputMaskBuilder or MultiOutputMaskBuilder.
+
+    Preserves backward compatibility for existing configs and code that rely
+    on the ``"sectioned"`` factory name.
+    """
+
+    def __init__(self):
+        self._single = SingleOutputMaskBuilder()
+        self._multi = MultiOutputMaskBuilder()
+
+    def build(self, item: dict, config, tokenizer) -> Optional[dict]:
+        sources_spec = getattr(config.input, "sources", None)
+        if sources_spec:
+            return self._multi.build(item, config, tokenizer)
+        return self._single.build(item, config, tokenizer)

astrai/preprocessing/packing.py

@@ -0,0 +1,121 @@
+"""Sequence packing strategies for shard-level reordering and truncation.
+
+Each strategy receives the accumulated ``{key: [list of token lists]}``
+dict for a shard and returns a reordered / truncated version.  The
+pipeline later flattens the result into contiguous tensors.
+"""
+
+from abc import ABC, abstractmethod
+from typing import Dict, List
+
+from astrai.factory import BaseFactory
+
+
+def _truncate(seq: List[int], max_len: int, mode: str) -> List[int]:
+    if len(seq) <= max_len:
+        return seq
+    if mode == "keep_end":
+        return seq[-max_len:]
+    return seq[:max_len]
+
+
+class PackingStrategy(ABC):
+    """Reorder and truncate sequences within a shard."""
+
+    @abstractmethod
+    def apply(
+        self,
+        keys: Dict[str, List[List[int]]],
+        max_packed_len: int,
+        truncation_mode: str,
+    ) -> Dict[str, List[List[int]]]:
+        raise NotImplementedError
+
+
+class PackingStrategyFactory(BaseFactory["PackingStrategy"]):
+    pass
+
+
+@PackingStrategyFactory.register("simple")
+class SimplePacking(PackingStrategy):
+    def apply(
+        self,
+        keys: Dict[str, List[List[int]]],
+        max_packed_len: int,
+        truncation_mode: str,
+    ) -> Dict[str, List[List[int]]]:
+        return {
+            k: [_truncate(v, max_packed_len, truncation_mode) for v in vals]
+            for k, vals in keys.items()
+        }
+
+
+@PackingStrategyFactory.register("bfd")
+class BFDPacking(PackingStrategy):
+    """Best-Fit Decreasing bin packing.
+
+    Assigns sequences to bins using a best-fit heuristic (sorted by
+    decreasing length) and concatenates sequences within each bin into
+    a single packed sequence.  Packed sequences are truncated to
+    *max_packed_len* so that each packed bin fits within one context
+    window during training.
+    """
+
+    def apply(
+        self,
+        keys: Dict[str, List[List[int]]],
+        max_packed_len: int,
+        truncation_mode: str,
+    ) -> Dict[str, List[List[int]]]:
+        sequences = keys.get("sequence", [])
+        if not sequences:
+            return keys
+        bins = self._plan(sequences, max_packed_len, truncation_mode)
+
+        packed: Dict[str, List[List[int]]] = {}
+        for k, vals in keys.items():
+            packed[k] = [
+                _truncate(
+                    self._concat_bin(vals, bin_indices),
+                    max_packed_len,
+                    truncation_mode,
+                )
+                for bin_indices in bins
+            ]
+        return packed
+
+    @staticmethod
+    def _concat_bin(vals: List[List[int]], indices: List[int]) -> List[int]:
+        result: List[int] = []
+        for i in indices:
+            result.extend(vals[i])
+        return result
+
+    @staticmethod
+    def _plan(
+        sequences: List[List[int]], max_packed_len: int, truncation_mode: str
+    ) -> List[List[int]]:
+        n = len(sequences)
+        order = sorted(range(n), key=lambda i: len(sequences[i]), reverse=True)
+        bins: List[List[int]] = []
+        bin_lengths: List[int] = []
+
+        for orig_idx in order:
+            seq_len = len(
+                _truncate(sequences[orig_idx], max_packed_len, truncation_mode)
+            )
+            best_bin = None
+            best_remain = max_packed_len + 1
+            for i, bl in enumerate(bin_lengths):
+                remain = max_packed_len - bl
+                if seq_len <= remain < best_remain:
+                    best_remain = remain
+                    best_bin = i
+            if best_bin is not None:
+                bins[best_bin].append(orig_idx)
+                bin_lengths[best_bin] += seq_len
+            else:
+                bins.append([orig_idx])
+                bin_lengths.append(seq_len)
+
+        return bins

astrai/preprocessing/pipeline.py

@@ -0,0 +1,192 @@
+"""Config-driven JSONL preprocessing pipeline.
+
+Composes a :class:`BaseMaskBuilder` (selected by ``input.type``) with
+sharding and flush to ``.h5`` / ``.bin`` storage.  Packing, position-id
+generation and storage writing are each delegated to pluggable strategies,
+dispatched by configuration keys.
+"""
+
+import json
+import logging
+import os
+from collections import defaultdict
+from itertools import chain
+from typing import Dict, List, Optional
+
+import torch
+import tqdm
+
+from astrai.config.preprocess_config import PipelineConfig
+from astrai.preprocessing.builder import MaskBuilderFactory
+from astrai.preprocessing.packing import PackingStrategyFactory
+from astrai.preprocessing.position_id import PositionIdStrategyFactory
+from astrai.preprocessing.writer import StoreWriterFactory
+from astrai.tokenize import AutoTokenizer
+
+logger = logging.getLogger(__name__)
+
+_STR_TO_DTYPE: dict[str, torch.dtype] = {
+    "bool": torch.bool,
+    "uint8": torch.uint8,
+    "int8": torch.int8,
+    "int16": torch.int16,
+    "int32": torch.int32,
+    "int64": torch.int64,
+    "float16": torch.float16,
+    "float32": torch.float32,
+    "float64": torch.float64,
+}
+
+
+def filter_by_length(text: str, min_len: int = 50, max_len: int = 2_000_000) -> bool:
+    return min_len <= len(text) <= max_len
+
+
+class Pipeline:
+    """Tokenization pipeline driven by a declarative :class:`PipelineConfig`.
+
+    Usage::
+
+        config = PipelineConfig.from_file("sft_pipeline.json")
+        Pipeline(config, ["data.jsonl"], output_dir="out", tokenizer_path="params").run()
+    """
+
+    def __init__(
+        self,
+        config: PipelineConfig,
+        input_paths: list[str],
+        output_dir: str,
+        tokenizer_path: str,
+    ):
+        os.makedirs(output_dir, exist_ok=True)
+        self.config = config
+        self.paths = input_paths
+        self.output_dir = output_dir
+        self.tokenizer_path = tokenizer_path
+
+        self.mask_builder = MaskBuilderFactory.create("sectioned")
+        self._packer = PackingStrategyFactory.create(
+            config.preprocessing.packing_strategy
+        )
+        self._position_id = PositionIdStrategyFactory.create(
+            config.output.position_ids_mode
+        )
+        self._writer = StoreWriterFactory.create(config.output.storage_format)
+
+    def transform(self, item: dict) -> Optional[dict]:
+        return self.mask_builder.build(item, self.config, self._tokenizer)
+
+    def run(self):
+        self._tokenizer = AutoTokenizer.from_pretrained(self.tokenizer_path)
+        domains: dict = defaultdict(lambda: defaultdict(list))
+        total_tokens = 0
+        shard_idx: dict[str, int] = defaultdict(int)
+        count = 0
+
+        pp = self.config.preprocessing
+
+        for item in tqdm.tqdm(
+            self._iter_items(), desc="Tokenizing", unit="docs", mininterval=0.5
+        ):
+            if pp.max_items and count >= pp.max_items:
+                break
+
+            try:
+                result = self.transform(item)
+            except Exception:
+                logger.warning(
+                    "Failed to process item #%d, skipping", count + 1, exc_info=True
+                )
+                continue
+            if result is None:
+                continue
+
+            domain = result.pop("domain", "__default__")
+
+            is_multi = bool(getattr(self.config.input, "sources", None))
+            if is_multi:
+                ids = self._primary_ids(result)
+            else:
+                ids = result.pop("sequence")
+                result["sequence"] = ids
+
+            if not ids:
+                continue
+
+            bucket = domains[domain]
+            self._align_bucket(bucket, result, ids)
+            for key, val in result.items():
+                bucket[key].append(val)
+
+            count += 1
+            total_tokens += len(ids)
+
+            if total_tokens >= self.config.output.max_tokens_per_shard:
+                self._flush(domains, shard_idx)
+                domains.clear()
+                total_tokens = 0
+
+        if total_tokens > 0:
+            self._flush(domains, shard_idx)
+
+    @staticmethod
+    def _primary_ids(result: dict) -> list:
+        """Return the first list-valued entry in *result* as the primary id
+        sequence for token counting."""
+        for val in result.values():
+            if isinstance(val, list) and val and isinstance(val[0], int):
+                return val
+        return []
+
+    @staticmethod
+    def _align_bucket(bucket: dict, result: dict, ids: list):
+        """Pad previously-accumulated keys that are missing from *result*."""
+        for key in list(bucket.keys()):
+            if key in result:
+                continue
+            bucket[key].append([0] * len(ids))
+
+    def _iter_items(self):
+        for path in self.paths:
+            with open(path, "r", encoding="utf-8") as f:
+                for line in f:
+                    line = line.strip()
+                    if not line:
+                        continue
+                    yield json.loads(line)
+
+    def _flush(self, domains, shard_idx):
+        for domain, keys in domains.items():
+            idx = shard_idx[domain]
+
+            pp = self.config.preprocessing
+            original_sequences = keys.get("sequence", [])
+            mode = self.config.output.position_ids_mode
+
+            if mode == "doc_reset" and original_sequences:
+                keys["position_ids"] = [list(range(len(s))) for s in original_sequences]
+
+            keys = self._packer.apply(dict(keys), pp.max_packed_len, pp.truncation_mode)
+
+            tensors: Dict[str, List[torch.Tensor]] = {}
+            for key, ids_list in keys.items():
+                dt = _STR_TO_DTYPE.get(
+                    self.config.output.dtype.get(key, "int32"), torch.int32
+                )
+                tensors[key] = [
+                    torch.tensor(list(chain.from_iterable(ids_list)), dtype=dt)
+                ]
+
+            if mode == "continuous" and original_sequences:
+                pos_ids = self._position_id.generate(keys.get("sequence", []))
+                if pos_ids:
+                    tensors["position_ids"] = [torch.tensor(pos_ids, dtype=torch.int32)]
+
+            self._writer.save(self.output_dir, domain, idx, tensors)
+            shard_idx[domain] = idx + 1
+
+            first_key = "sequence" if "sequence" in tensors else next(iter(tensors))
+            tqdm.tqdm.write(
+                f"  saved {domain}/shard_{idx:04d}  "
+                f"({tensors[first_key][0].numel():,} tokens)"
+            )

astrai/preprocessing/position_id.py

@@ -0,0 +1,46 @@
+"""Position-id generation strategies for packed sequences.
+
+Each strategy takes the list of per-document token sequences after packing
+and returns a flat list of position ids (same total length as all
+sequences combined).  The pipeline wraps the result into a tensor and
+attaches it as ``position_ids``.
+"""
+
+from abc import ABC, abstractmethod
+from typing import List
+
+from astrai.factory import BaseFactory
+
+
+class PositionIdStrategy(ABC):
+    """Generate ``position_ids`` for packed sequences."""
+
+    @abstractmethod
+    def generate(self, sequences: List[List[int]]) -> List[int]:
+        raise NotImplementedError
+
+
+class PositionIdStrategyFactory(BaseFactory["PositionIdStrategy"]):
+    pass
+
+
+@PositionIdStrategyFactory.register("none")
+class NoPositionId(PositionIdStrategy):
+    def generate(self, sequences: List[List[int]]) -> List[int]:
+        return []
+
+
+@PositionIdStrategyFactory.register("doc_reset")
+class DocResetPositionId(PositionIdStrategy):
+    def generate(self, sequences: List[List[int]]) -> List[int]:
+        pos_ids = []
+        for seq in sequences:
+            pos_ids.extend(range(len(seq)))
+        return pos_ids
+
+
+@PositionIdStrategyFactory.register("continuous")
+class ContinuousPositionId(PositionIdStrategy):
+    def generate(self, sequences: List[List[int]]) -> List[int]:
+        total = sum(len(seq) for seq in sequences)
+        return list(range(total))

astrai/preprocessing/writer.py

@@ -0,0 +1,75 @@
+"""Storage writer strategies for pipeline output.
+
+The :class:`StoreWriter` abstraction decouples the pipeline from the
+concrete storage format (bin / h5).  The pipeline builds a ``{key:
+List[Tensor]}`` dict and delegates the write to the writer selected
+by ``output.storage_format``.
+"""
+
+import logging
+import os
+import shutil
+from abc import ABC, abstractmethod
+from typing import Dict, List
+
+import torch
+
+from astrai.factory import BaseFactory
+from astrai.serialization import save_bin, save_h5
+
+logger = logging.getLogger(__name__)
+
+
+class StoreWriter(ABC):
+    """Write pre-tokenized tensors to disk in a format-specific way."""
+
+    @abstractmethod
+    def save(
+        self,
+        output_dir: str,
+        domain: str,
+        shard_idx: int,
+        tensors: Dict[str, List[torch.Tensor]],
+    ) -> None: ...
+
+
+class StoreWriterFactory(BaseFactory["StoreWriter"]):
+    pass
+
+
+@StoreWriterFactory.register("bin")
+class BinWriter(StoreWriter):
+    def save(self, output_dir, domain, shard_idx, tensors):
+        shard_path = os.path.join(output_dir, domain, f"shard_{shard_idx:04d}")
+        try:
+            save_bin(shard_path, tensors)
+        except Exception:
+            if os.path.exists(shard_path):
+                shutil.rmtree(shard_path, ignore_errors=True)
+            logger.error(
+                "Failed to write shard %s/%s_%04d, cleaned up partial output",
+                domain,
+                "shard",
+                shard_idx,
+                exc_info=True,
+            )
+            raise
+
+
+@StoreWriterFactory.register("h5")
+class H5Writer(StoreWriter):
+    def save(self, output_dir, domain, shard_idx, tensors):
+        chunk_dir = os.path.join(output_dir, domain)
+        file_path = os.path.join(chunk_dir, f"data_{shard_idx:04d}.h5")
+        try:
+            save_h5(chunk_dir, f"data_{shard_idx:04d}", tensors)
+        except Exception:
+            if os.path.exists(file_path):
+                os.remove(file_path)
+            logger.error(
+                "Failed to write shard %s/data_%04d.h5, cleaned up partial output",
+                domain,
+                shard_idx,
+                exc_info=True,
+            )
+            raise

astrai/serialization/__init__.py

@@ -0,0 +1,43 @@
+"""Serialization utilities for models and datasets.
+
+This package re-exports checkpoint helpers and dataset storage helpers so
+that existing imports from ``astrai.serialization`` continue to work.
+"""
+
+from astrai.serialization.checkpoint import (
+    Checkpoint,
+    load_json,
+    load_model_config,
+    load_model_weights,
+    load_safetensors,
+    load_state_dict,
+    load_torch,
+    save_json,
+    save_model,
+    save_safetensors,
+    save_torch,
+)
+from astrai.serialization.dataset import (
+    load_bin,
+    load_h5,
+    save_bin,
+    save_h5,
+)
+
+__all__ = [
+    "Checkpoint",
+    "load_json",
+    "load_model_config",
+    "load_model_weights",
+    "load_safetensors",
+    "load_state_dict",
+    "load_torch",
+    "save_json",
+    "save_model",
+    "save_safetensors",
+    "save_torch",
+    "load_bin",
+    "load_h5",
+    "save_bin",
+    "save_h5",
+]

astrai/serialization/checkpoint.py

@@ -1,9 +1,12 @@
+"""Model checkpoint serialization helpers."""
+
 import io
 import json
+import os
 import time
 from dataclasses import dataclass, field
 from pathlib import Path
-from typing import Any, Dict, Union
+from typing import Any, Dict, Optional, Union
 
 import safetensors.torch as st
 import torch
@@ -136,7 +139,7 @@ def load_state_dict(path: Union[str, Path], broadcast: bool = False) -> dict:
 class Checkpoint:
     state_dict: Dict[str, Any] = field(default_factory=dict)
     epoch: int = 0
-    iteration: int = 0
+    consumed_samples: int = 0
     extra: Dict[str, Any] = field(default_factory=dict)
     meta: Dict[str, Any] = field(default_factory=dict)
     config: Dict[str, Any] = field(default_factory=dict)
@@ -150,7 +153,7 @@ class Checkpoint:
 
         meta = {
             "epoch": self.epoch,
-            "iteration": self.iteration,
+            "consumed_samples": self.consumed_samples,
             "timestamp": time.strftime("%Y-%m-%dT%H:%M:%S"),
             **self.meta,
         }
@@ -176,7 +179,26 @@ class Checkpoint:
         return cls(
             state_dict=state_dict,
             epoch=meta.get("epoch", 0),
-            iteration=meta.get("iteration", 0),
+            consumed_samples=meta.get("consumed_samples", 0),
             extra=extra,
             config=config,
         )
+
+    @classmethod
+    def load_any(cls, save_dir: str, broadcast: bool = False) -> Optional["Checkpoint"]:
+        save_path = Path(save_dir)
+        meta_path = save_path / _META_FILE
+        weights_path = save_path / _WEIGHTS_FILE
+
+        if meta_path.exists():
+            return cls.load(save_dir, broadcast=broadcast)
+
+        if weights_path.exists():
+            state_dict = load_state_dict(weights_path, broadcast=broadcast)
+            config = {}
+            config_path = save_path / _CONFIG_FILE
+            if config_path.exists():
+                config = load_json(config_path, broadcast)
+            return cls(state_dict=state_dict, config=config)
+
+        return None

astrai/serialization/dataset.py

@@ -0,0 +1,76 @@
+"""Dataset storage serialization helpers (HDF5 / memory-mapped binary)."""
+
+import json
+import os
+from pathlib import Path
+from typing import Dict, List
+
+import h5py
+import numpy as np
+import torch
+from torch import Tensor
+
+
+def save_h5(file_path: str, file_name: str, tensor_group: Dict[str, List[Tensor]]):
+    os.makedirs(file_path, exist_ok=True)
+    full_file_path = os.path.join(file_path, f"{file_name}.h5")
+    with h5py.File(full_file_path, "w") as f:
+        for key, tensors in tensor_group.items():
+            grp = f.create_group(key)
+            for idx, tensor in enumerate(tensors):
+                arr = tensor.cpu().numpy()
+                grp.create_dataset(f"data_{idx}", data=arr)
+
+
+def load_h5(file_path: str, share_memory=True) -> Dict[str, List[Tensor]]:
+    tensor_group: Dict[str, List[Tensor]] = {}
+
+    root_path = Path(file_path)
+    if root_path.is_file() and root_path.suffix in (".h5", ".hdf5"):
+        h5_files = [root_path]
+    else:
+        h5_files = list(root_path.rglob("*.h5")) + list(root_path.rglob("*.hdf5"))
+
+    for h5_file in h5_files:
+        with h5py.File(h5_file, "r") as f:
+            for key in f.keys():
+                grp = f[key]
+                dsets = []
+                for dset_name in grp.keys():
+                    dset = grp[dset_name]
+                    tensor = torch.from_numpy(dset[:])
+                    if share_memory:
+                        tensor = tensor.share_memory_()
+                    dsets.append(tensor)
+
+                if tensor_group.get(key) is None:
+                    tensor_group[key] = []
+                tensor_group[key].extend(dsets)
+
+    return tensor_group
+
+
+def save_bin(file_path: str, tensor_group: Dict[str, List[Tensor]]):
+    os.makedirs(file_path, exist_ok=True)
+    meta = {}
+    for key, tensors in tensor_group.items():
+        cat = torch.cat(tensors, dim=0)
+        meta[key] = {"shape": list(cat.shape), "dtype": str(cat.dtype).split(".")[-1]}
+        np.asarray(cat.cpu().numpy()).tofile(os.path.join(file_path, f"{key}.bin"))
+    with open(os.path.join(file_path, "meta.json"), "w") as f:
+        json.dump(meta, f)
+
+
+def load_bin(file_path: str) -> Dict[str, List[Tensor]]:
+    with open(os.path.join(file_path, "meta.json"), "r") as f:
+        meta = json.load(f)
+    segments: Dict[str, List[Tensor]] = {}
+    for key, info in meta.items():
+        arr = np.memmap(
+            os.path.join(file_path, f"{key}.bin"),
+            dtype=info["dtype"],
+            mode="r+",
+            shape=tuple(info["shape"]),
+        )
+        segments[key] = [torch.from_numpy(arr)]
+    return segments

astrai/tokenize/chat_template.py

@@ -1,13 +1,10 @@
-from dataclasses import dataclass
 from typing import Any, Dict, List, Optional
 
 from jinja2 import Template
 
-# Message type for chat messages
 type MessageType = Dict[str, Any]
 
 
-@dataclass
 class ChatTemplate:
     """A chat template with Jinja2 rendering support.
 
@@ -15,23 +12,24 @@ class ChatTemplate:
         name: Unique identifier for the template.
         template_str: Jinja2 template string.
         description: Optional description.
-        default_variables: Optional dictionary of default variable values
-            that will be passed to the template if not overridden during rendering.
+        default_variables: Optional dictionary of default variable values.
         special_tokens: Optional dictionary mapping token names to their string values.
-            These tokens are automatically added to the template variables.
     """
 
-    name: str
-    template_str: str
-    description: str = ""
-    default_variables: Dict[str, Any] = None
-    special_tokens: Dict[str, str] = None
-
-    def __post_init__(self):
-        if self.default_variables is None:
-            self.default_variables = {}
-        if self.special_tokens is None:
-            self.special_tokens = {}
+    def __init__(
+        self,
+        name: str = "",
+        template_str: str = "",
+        description: str = "",
+        default_variables: Optional[Dict[str, Any]] = None,
+        special_tokens: Optional[Dict[str, str]] = None,
+    ):
+        self.name = name
+        self.template_str = template_str
+        self.description = description
+        self.default_variables = default_variables or {}
+        self.special_tokens = special_tokens or {}
+        self._compiled: Template = Template(template_str)
 
     @classmethod
     def from_string(
@@ -43,7 +41,7 @@ class ChatTemplate:
     ) -> "ChatTemplate":
         """Create a ChatTemplate instance directly from a template string."""
         return cls(
-            name="",  # empty name for ad‑hoc templates
+            name="",
             template_str=template_str,
             description=description,
             default_variables=default_variables,
@@ -73,5 +71,4 @@ class ChatTemplate:
         if system_prompt is not None:
             variables["system_prompt"] = system_prompt
 
-        jinja_template = Template(self.template_str)
-        return jinja_template.render(**variables)
+        return self._compiled.render(**variables)

astrai/trainer/__init__.py

@@ -1,4 +1,3 @@
-from astrai.trainer.optim import Muon
 from astrai.trainer.schedule import BaseScheduler, SchedulerFactory
 from astrai.trainer.strategy import BaseStrategy, StrategyFactory
 from astrai.trainer.train_callback import (
@@ -10,8 +9,6 @@ from astrai.trainer.trainer import Trainer
 __all__ = [
     # Main trainer
     "Trainer",
-    # Optimizer
-    "Muon",
     # Strategy factory
     "StrategyFactory",
     "BaseStrategy",

astrai/trainer/metric_util.py

@@ -1,42 +1,25 @@
-from typing import Any, Callable, Dict
+from typing import Dict
 
 import torch
 import torch.nn as nn
 
 
-def _grad_stat(
-    model: nn.Module, fn: Callable[[torch.Tensor], Any], default: Any
-) -> dict:
-    results = {}
-    for name, param in model.named_parameters():
-        results[name] = default
-        if param.grad is not None:
-            results[name] = fn(param.grad.data)
-    return results
+def grad_norm(model: nn.Module, per_param: bool = False) -> float | Dict[str, float]:
+    grads = [p.grad.detach() for p in model.parameters() if p.grad is not None]
+    if not grads:
+        return 0.0
 
-
-def grad_norm(model: nn.Module, norm_type: int = 2) -> Dict[str, float]:
-    return _grad_stat(model, lambda g: g.norm(norm_type).item(), 0.0)
-
-
-def grad_std(model: nn.Module) -> Dict[str, float]:
-    return _grad_stat(model, lambda g: g.std().item(), 0.0)
-
-
-def grad_max(model: nn.Module) -> Dict[str, float]:
-    return _grad_stat(model, lambda g: g.max().item(), -float("inf"))
-
-
-def grad_min(model: nn.Module) -> Dict[str, float]:
-    return _grad_stat(model, lambda g: g.min().item(), float("inf"))
-
-
-def grad_mean(model: nn.Module) -> Dict[str, float]:
-    return _grad_stat(model, lambda g: g.mean().item(), 0.0)
-
-
-def grad_nan_num(model: nn.Module) -> Dict[str, int]:
-    return _grad_stat(model, lambda g: g.isnan().sum().item(), 0)
+    total_sq = torch.stack([g.pow(2).sum() for g in grads]).sum()
+    if per_param:
+        norms = {}
+        for name, param in model.named_parameters():
+            if param.grad is not None:
+                norms[name] = param.grad.norm(2).item()
+            else:
+                norms[name] = 0.0
+        norms["total"] = total_sq.sqrt().item()
+        return norms
+    return total_sq.sqrt().item()
 
 
 def ctx_get_loss(ctx):
@@ -52,24 +35,4 @@ def ctx_get_val_loss(ctx):
 
 
 def ctx_get_grad_norm(ctx):
-    return grad_norm(ctx.model)
-
-
-def ctx_get_grad_std(ctx):
-    return grad_std(ctx.model)
-
-
-def ctx_get_grad_max(ctx):
-    return grad_max(ctx.model)
-
-
-def ctx_get_grad_min(ctx):
-    return grad_min(ctx.model)
-
-
-def ctx_get_grad_mean(ctx):
-    return grad_mean(ctx.model)
-
-
-def ctx_get_grad_nan_num(ctx):
-    return grad_nan_num(ctx.model)
+    return ctx.grad_norm

astrai/trainer/optim.py

@@ -1,143 +0,0 @@
-import torch
-from torch.optim import Optimizer
-
-
-def _zeropower_via_newtonschulz(G: torch.Tensor, steps: int = 5):
-    assert G.ndim == 2
-    X = G
-    scale = max(1, G.size(0) / G.size(1)) ** 0.5
-    X = X / (X.norm() + 1e-7) * scale
-    if steps == 0:
-        return X
-    a, b, c = (3.4445, -4.7750, 2.0315)
-    for _ in range(steps):
-        A = X @ X.T
-        B = A @ X
-        X = a * X + b * B + c * (A @ B)
-    return X
-
-
-class Muon(Optimizer):
-    def __init__(
-        self,
-        params,
-        lr: float = 2e-3,
-        momentum: float = 0.95,
-        weight_decay: float = 0.0,
-        nesterov: bool = True,
-        ns_steps: int = 5,
-        adamw_lr: float = None,
-        adamw_betas: tuple = (0.9, 0.95),
-        adamw_eps: float = 1e-8,
-        adamw_wd: float = 0.0,
-    ):
-        defaults = dict(
-            lr=lr,
-            momentum=momentum,
-            weight_decay=weight_decay,
-            nesterov=nesterov,
-            ns_steps=ns_steps,
-            adamw_lr=adamw_lr if adamw_lr is not None else lr * 0.1,
-            adamw_betas=adamw_betas,
-            adamw_eps=adamw_eps,
-            adamw_wd=adamw_wd,
-        )
-        super().__init__(params, defaults)
-
-    @torch.no_grad()
-    def step(self, closure=None):
-        loss = None
-        if closure is not None:
-            with torch.enable_grad():
-                loss = closure()
-
-        for group in self.param_groups:
-            params_2d, params_1d = [], []
-            grads_2d, grads_1d = [], []
-
-            for p in group["params"]:
-                if p.grad is None:
-                    continue
-                if p.grad.is_sparse:
-                    raise RuntimeError("Muon does not support sparse gradients")
-                if p.ndim >= 2:
-                    params_2d.append(p)
-                    grads_2d.append(p.grad)
-                else:
-                    params_1d.append(p)
-                    grads_1d.append(p.grad)
-
-            if params_2d:
-                self._muon_update_foreach(params_2d, grads_2d, group)
-            if params_1d:
-                self._adamw_update_foreach(params_1d, grads_1d, group)
-
-        return loss
-
-    def _muon_update_foreach(self, params_2d, grads_2d, group):
-        lr = group["lr"]
-        momentum = group["momentum"]
-        wd = group["weight_decay"]
-        nesterov = group["nesterov"]
-        ns_steps = group["ns_steps"]
-
-        if wd != 0:
-            torch._foreach_mul_(params_2d, 1 - lr * wd)
-
-        if nesterov:
-            grads_2d = torch._foreach_add(grads_2d, params_2d, alpha=wd)
-
-        bufs = []
-        for p, grad in zip(params_2d, grads_2d):
-            state = self.state[p]
-            if "momentum_buffer" not in state:
-                state["momentum_buffer"] = torch.zeros_like(grad)
-            bufs.append(state["momentum_buffer"])
-
-        torch._foreach_lerp_(bufs, grads_2d, 1 - momentum)
-
-        for p, buf in zip(params_2d, bufs):
-            update = _zeropower_via_newtonschulz(buf, steps=ns_steps)
-            scale = max(1, p.size(0) / p.size(1)) ** 0.5
-            p.add_(update, alpha=-lr * scale)
-
-    def _adamw_update_foreach(self, params_1d, grads_1d, group):
-        lr = group["adamw_lr"]
-        betas = group["adamw_betas"]
-        eps = group["adamw_eps"]
-        wd = group["adamw_wd"]
-
-        steps: list[int] = []
-        exp_avgs, exp_avg_sqs = [], []
-        has_state = []
-        for p in params_1d:
-            state = self.state[p]
-            if not state:
-                state["step"] = 0
-                state["exp_avg"] = torch.zeros_like(p)
-                state["exp_avg_sq"] = torch.zeros_like(p)
-                has_state.append(False)
-            else:
-                has_state.append(True)
-            state["step"] += 1
-            steps.append(state["step"])
-            exp_avgs.append(state["exp_avg"])
-            exp_avg_sqs.append(state["exp_avg_sq"])
-
-        beta1, beta2 = betas
-
-        torch._foreach_lerp_(exp_avgs, grads_1d, 1 - beta1)
-        grads_sq = torch._foreach_mul(grads_1d, grads_1d)
-        torch._foreach_lerp_(exp_avg_sqs, grads_sq, 1 - beta2)
-
-        bias_correction1 = [1 - beta1**s for s in steps]
-        bias_correction2 = [1 - beta2**s for s in steps]
-
-        if wd != 0:
-            torch._foreach_mul_(params_1d, 1 - lr * wd)
-
-        exp_avg_corrected = torch._foreach_div(exp_avgs, bias_correction1)
-        denom = torch._foreach_div(exp_avg_sqs, bias_correction2)
-        denom = torch._foreach_sqrt(denom)
-        torch._foreach_add_(denom, eps)
-        torch._foreach_addcdiv_(params_1d, exp_avg_corrected, denom, value=-lr)

astrai/trainer/schedule.py

@@ -2,7 +2,7 @@
 
 import math
 from abc import ABC, abstractmethod
-from typing import Any, Dict, List, Type
+from typing import Any, Dict, List
 
 from torch.optim.lr_scheduler import LRScheduler
 
@@ -31,7 +31,6 @@ class SchedulerFactory(BaseFactory["BaseScheduler"]):
     """Factory class for creating learning rate schedulers.
 
     Supports decorator-based registration for extensible scheduler types.
-    Also supports creation from ScheduleConfig objects.
 
     Example usage:
         @SchedulerFactory.register("custom")
@@ -41,33 +40,6 @@ class SchedulerFactory(BaseFactory["BaseScheduler"]):
         scheduler = SchedulerFactory.create("custom", optimizer, **kwargs)
     """
 
-    @classmethod
-    def _validate_component(cls, scheduler_cls: Type[BaseScheduler]):
-        """Validate that the scheduler class inherits from BaseScheduler."""
-        if not issubclass(scheduler_cls, BaseScheduler):
-            raise TypeError(f"{scheduler_cls.__name__} must inherit from BaseScheduler")
-
-    @classmethod
-    def create(
-        cls, optimizer, schedule_type: str = "none", **kwargs
-    ) -> "BaseScheduler":
-        """Create a scheduler instance by type name.
-
-        Args:
-            optimizer: PyTorch optimizer
-            schedule_type: Type of scheduler ("cosine", "sgdr")
-            **kwargs: Arguments passed to the scheduler constructor
-
-        Returns:
-            Scheduler instance
-        """
-        return super().create(schedule_type, optimizer, **kwargs)
-
-    @classmethod
-    def available_types(cls) -> list:
-        """Return list of registered scheduler type names."""
-        return cls.list_registered()
-
 
 # ----------- Scheduler implementations -----------
 
@@ -81,7 +53,7 @@ class CosineScheduler(BaseScheduler):
         optimizer,
         warmup_steps: int,
         lr_decay_steps: int,
-        min_rate: float = 0.05,
+        min_rate: float = 0.01,
         last_epoch: int = -1,
     ):
         self.warmup_steps = warmup_steps
@@ -93,11 +65,15 @@ class CosineScheduler(BaseScheduler):
     def get_lr(self) -> List[float]:
         # warmup
         if self.last_epoch < self.warmup_steps:
-            warmup_factor = max(self.min_rate, self.last_epoch / self.warmup_steps)
+            warmup_factor = max(
+                self.min_rate, self.last_epoch / max(self.warmup_steps, 1)
+            )
             return [base_lr * warmup_factor for base_lr in self.base_lrs]
 
         # cosine decay
-        decay_progress = (self.last_epoch - self.warmup_steps) / self.lr_decay_steps
+        decay_progress = (self.last_epoch - self.warmup_steps) / max(
+            self.lr_decay_steps, 1
+        )
         decay_progress = min(decay_progress, 1.0)
         cosine_decay = 0.5 * (1.0 + math.cos(math.pi * decay_progress))
         decay_factor = max(self.min_rate, cosine_decay)
@@ -132,7 +108,7 @@ class SGDRScheduler(BaseScheduler):
         optimizer,
         warmup_steps: int,
         cycle_length: int,
-        min_rate: float = 0.05,
+        min_rate: float = 0.01,
         t_mult: int = 2,
         last_epoch: int = -1,
     ):
@@ -146,7 +122,9 @@ class SGDRScheduler(BaseScheduler):
     def get_lr(self):
         # warmup
         if self.last_epoch < self.warmup_steps:
-            warmup_factor = max(self.min_rate, self.last_epoch / self.warmup_steps)
+            warmup_factor = max(
+                self.min_rate, self.last_epoch / max(self.warmup_steps, 1)
+            )
             return [base_lr * warmup_factor for base_lr in self.base_lrs]
 
         # SGDR
@@ -192,3 +170,66 @@ class SGDRScheduler(BaseScheduler):
         self.min_rate = state_dict.pop("min_rate")
         self.t_mult = state_dict.pop("t_mult")
         super().load_state_dict(state_dict)
+
+
+@SchedulerFactory.register("wsd")
+class WSDScheduler(BaseScheduler):
+    """WSD (Warmup-Stable-Decay) scheduler with sqrt cooldown.
+
+    warmup_steps:  linear warmup from min_rate to 1.0
+    stable_steps:   constant at base_lr
+    decay_steps:    sqrt decay from base_lr to min_rate
+    min_rate:       minimum lr as fraction of base_lr (default 0.0)
+    """
+
+    def __init__(
+        self,
+        optimizer,
+        warmup_steps: int,
+        stable_steps: int,
+        decay_steps: int,
+        min_rate: float = 0.01,
+        last_epoch: int = -1,
+    ):
+        self.warmup_steps = warmup_steps
+        self.stable_steps = stable_steps
+        self.decay_steps = decay_steps
+        self.min_rate = min_rate
+        self.total_steps = warmup_steps + stable_steps + decay_steps
+        super().__init__(optimizer, last_epoch)
+
+    def get_lr(self) -> List[float]:
+        if self.last_epoch < self.warmup_steps:
+            factor = max(self.min_rate, self.last_epoch / max(self.warmup_steps, 1))
+            return [base_lr * factor for base_lr in self.base_lrs]
+
+        offset = self.last_epoch - self.warmup_steps
+
+        if offset < self.stable_steps:
+            return list(self.base_lrs)
+
+        decay_ratio = (offset - self.stable_steps) / max(self.decay_steps, 1)
+        decay_ratio = min(decay_ratio, 1.0)
+        factor = (1.0 - self.min_rate) * (1.0 - decay_ratio) ** 2 + self.min_rate
+        return [base_lr * factor for base_lr in self.base_lrs]
+
+    def state_dict(self):
+        state = super().state_dict()
+        state.update(
+            {
+                "warmup_steps": self.warmup_steps,
+                "stable_steps": self.stable_steps,
+                "decay_steps": self.decay_steps,
+                "min_rate": self.min_rate,
+                "total_steps": self.total_steps,
+            }
+        )
+        return state
+
+    def load_state_dict(self, state_dict):
+        self.warmup_steps = state_dict.pop("warmup_steps")
+        self.stable_steps = state_dict.pop("stable_steps")
+        self.decay_steps = state_dict.pop("decay_steps")
+        self.min_rate = state_dict.pop("min_rate")
+        self.total_steps = state_dict.pop("total_steps")
+        super().load_state_dict(state_dict)

astrai/trainer/strategy.py

@@ -1,41 +1,28 @@
 """Training strategy implementations with factory pattern."""
 
-import copy
 from abc import ABC, abstractmethod
-from typing import Any, Callable, Dict, Union
+from typing import Callable, Dict, Union
 
 import torch
 import torch.nn as nn
 import torch.nn.functional as F
 from torch import Tensor
-from torch.distributed.fsdp import FullyShardedDataParallel as FSDP
-from torch.nn.parallel import DistributedDataParallel as DDP
 
 from astrai.factory import BaseFactory
 
 
-def unwrap_model(model: nn.Module) -> nn.Module:
-    if isinstance(model, DDP):
-        return model.module
-    if isinstance(model, FSDP):
-        return model._fsdp_wrapped_module
-    return model
-
-
-def create_ref_model(model: nn.Module) -> nn.Module:
-    """Create a reference model for DPO/GRPO training.
-
-    Handles DDP-wrapped models safely by unwrapping first,
-    then creating a deep copy with frozen gradients.
-    """
-    original_model = unwrap_model(model)
-    ref_model = copy.deepcopy(original_model)
+def create_ref_model(
+    model_fn: Callable[[], nn.Module], state_dict: Dict[str, Tensor]
+) -> nn.Module:
+    """Create a frozen reference model from model_fn + full state dict."""
+    ref_model = model_fn()
+    ref_model.load_state_dict(state_dict)
     ref_model.requires_grad_(False)
     ref_model.eval()
     return ref_model
 
 
-def move_to_device(batch: Dict[str, Tensor], device: str) -> Any:
+def move_to_device(batch: Dict[str, Tensor], device: str) -> Dict[str, Tensor]:
     """Move batch tensors to specified device with non-blocking transfer."""
     return {key: value.to(device, non_blocking=True) for key, value in batch.items()}
 
@@ -45,7 +32,7 @@ def get_logprobs(
     input_ids: Tensor,
     mask: Tensor,
     reduction: str,
-):
+) -> Tensor:
     """Compute token-wise log probabilities from model outputs.
 
     Args:
@@ -83,14 +70,35 @@ def get_logprobs(
         return token_logprobs * shifted_mask
 
 
+def make_doc_boundary_mask(position_ids: Tensor) -> Tensor:
+    S = position_ids.size(1)
+    device = position_ids.device
+    boundaries = position_ids[:, 1:] <= position_ids[:, :-1]
+    doc_ids = torch.cat(
+        [
+            torch.zeros(position_ids.size(0), 1, dtype=torch.long, device=device),
+            boundaries.long().cumsum(dim=1),
+        ],
+        dim=1,
+    )
+    same_doc = doc_ids.unsqueeze(-1) == doc_ids.unsqueeze(-2)
+    causal = torch.tril(torch.ones(S, S, dtype=torch.bool, device=device))
+    return (same_doc & causal).unsqueeze(1)
+
+
 class BaseStrategy(ABC):
     """Abstract base class for training strategies."""
 
     def __init__(
-        self, model: Union[Callable[..., Dict[str, Tensor]]], device: str, **kwargs
+        self,
+        model: Union[nn.Module, Callable[..., Dict[str, Tensor]]],
+        device: str,
+        **kwargs,
     ):
         self.model = model
         self.device = device
+        self.executor = kwargs.pop("executor", None)
+        self.model_fn = kwargs.pop("model_fn", None)
         self.extra_kwargs = kwargs
 
     @abstractmethod
@@ -124,32 +132,6 @@ class StrategyFactory(BaseFactory["BaseStrategy"]):
         strategy = StrategyFactory.create("custom", model, device)
     """
 
-    @classmethod
-    def _validate_component(cls, strategy_cls: type):
-        """Validate that the strategy class inherits from BaseStrategy."""
-        if not issubclass(strategy_cls, BaseStrategy):
-            raise TypeError(f"{strategy_cls.__name__} must inherit from BaseStrategy")
-
-    @classmethod
-    def create(cls, train_type: str, model, device: str, **kwargs) -> "BaseStrategy":
-        """Create a strategy instance based on training type.
-
-        Args:
-            train_type: Type of training ("seq", "sft", "dpo", "grpo")
-            model: Model instance for the strategy
-            device: Device to run the strategy on
-            **kwargs: Additional arguments passed to strategy constructor
-
-        Returns:
-            Strategy instance
-        """
-        return super().create(train_type, model, device, **kwargs)
-
-    @classmethod
-    def available_strategies(cls) -> list:
-        """Return list of registered strategy names."""
-        return cls.list_registered()
-
 
 # ============== Strategy Classes ==============
 # All strategies are registered at class definition time using the decorator
@@ -162,7 +144,13 @@ class SEQStrategy(BaseStrategy):
     Computes cross-entropy loss for next token prediction.
     """
 
-    def __init__(self, model, device, label_smoothing: float = 0.0, **kwargs):
+    def __init__(
+        self,
+        model: Union[nn.Module, Callable[..., Dict[str, Tensor]]],
+        device: str,
+        label_smoothing: float = 0.0,
+        **kwargs,
+    ):
         super().__init__(model, device, **kwargs)
         self.label_smoothing = label_smoothing
 
@@ -187,21 +175,31 @@ class SFTStrategy(BaseStrategy):
     Applies cross-entropy loss only to tokens where loss_mask is True.
     """
 
-    def __init__(self, model, device, label_smoothing: float = 0.0, **kwargs):
+    def __init__(
+        self,
+        model: Union[nn.Module, Callable[..., Dict[str, Tensor]]],
+        device: str,
+        label_smoothing: float = 0.0,
+        **kwargs,
+    ):
         super().__init__(model, device, **kwargs)
         self.label_smoothing = label_smoothing
 
     def compute_loss(self, batch: Dict[str, Tensor]) -> Tensor:
         batch = move_to_device(batch, self.device)
-        input_ids, target_ids, loss_mask = (
+        input_ids, target_ids, position_ids, loss_mask = (
             batch["input_ids"],
             batch["target_ids"],
+            batch["position_ids"],
             batch["loss_mask"],
         )
 
         ignore_index = -100
-        logits = self.model(input_ids=input_ids)["logits"]
-        target_ids = target_ids.masked_fill(loss_mask == 0, ignore_index)
+        input_mask = make_doc_boundary_mask(position_ids)
+        target_ids = target_ids.masked_fill(~loss_mask, ignore_index)
+        logits = self.model(
+            input_ids=input_ids, position_ids=position_ids, input_mask=input_mask
+        )["logits"]
 
         loss = F.cross_entropy(
             input=logits.flatten(0, 1).float(),
@@ -230,7 +228,9 @@ class DPOStrategy(BaseStrategy):
         **kwargs,
     ):
         super().__init__(model, device, **kwargs)
-        self.ref_model = create_ref_model(model)
+        self.ref_model = create_ref_model(
+            self.model_fn, self.executor.unwrap_model(model)
+        ).to(device=self.device)
         self.beta = beta
         self.reduction = reduction
 
@@ -284,7 +284,9 @@ class GRPOStrategy(BaseStrategy):
         **kwargs,
     ):
         super().__init__(model, device, **kwargs)
-        self.ref_model = create_ref_model(model)
+        self.ref_model = create_ref_model(
+            self.model_fn, self.executor.unwrap_model(model)
+        ).to(device=self.device)
         self.clip_eps = clip_eps
         self.kl_coef = kl_coef
         self.group_size = group_size
@@ -294,8 +296,7 @@ class GRPOStrategy(BaseStrategy):
 
     def sync_ref_model(self):
         """Copy current model weights to ref model."""
-        ref_state = self.model.state_dict()
-        self.ref_model.load_state_dict(ref_state)
+        self.ref_model.load_state_dict(self.executor.unwrap_model(self.model))
 
     def compute_loss(self, batch: Dict[str, Tensor]) -> Tensor:
         self._step += 1

astrai/trainer/train_callback.py

@@ -9,7 +9,6 @@ from typing import IO, Callable, List, Optional, Protocol, runtime_checkable
 import torch
 import torch.distributed as dist
 import torch.nn as nn
-from torch.nn.utils import clip_grad_norm_
 from torch.utils.checkpoint import checkpoint as torch_checkpoint
 from tqdm import tqdm
 
@@ -18,12 +17,7 @@ from astrai.parallel import only_on_rank
 from astrai.parallel.setup import get_current_device, get_rank
 from astrai.serialization import Checkpoint
 from astrai.trainer.metric_util import (
-    ctx_get_grad_max,
-    ctx_get_grad_mean,
-    ctx_get_grad_min,
-    ctx_get_grad_nan_num,
     ctx_get_grad_norm,
-    ctx_get_grad_std,
     ctx_get_loss,
     ctx_get_lr,
     ctx_get_val_loss,
@@ -86,7 +80,9 @@ class GradientClippingCallback(TrainCallback):
         self.max_grad_norm = max_grad_norm
 
     def on_optimizer_step(self, context: TrainContext):
-        clip_grad_norm_(context.model.parameters(), self.max_grad_norm)
+        context.grad_norm = context.executor.clip_grad_norm(
+            context.model, self.max_grad_norm
+        )
 
 
 @CallbackFactory.register("gradient_checkpointing")
@@ -143,33 +139,35 @@ class CheckpointCallback(TrainCallback):
         self.interval = interval
         self.weight_only = weight_only
         self.save_extra_fn = save_extra_fn or CheckpointCallback.save_extra
-        self.last_ckpt_iter = 0
+        self.last_ckpt_step = 0
 
     def _save_checkpoint(self, context: TrainContext):
-        unwrapped = context.executor.unwrap_model(context.model)
-        state_dict = unwrapped.state_dict()
-        self.last_ckpt_iter = context.iteration
+        state_dict = context.executor.unwrap_model(context.model)
+        self.last_ckpt_step = context.optimizer_step
 
         if get_rank() == 0:
             save_path = os.path.join(
-                self.save_dir, f"epoch_{context.epoch}_iter_{context.iteration}"
+                self.save_dir,
+                f"epoch_{context.epoch}_step_{context.optimizer_step}",
             )
             extra = self.save_extra_fn(context)
+            meta = context.config.to_dict()
             context.checkpoint = Checkpoint(
                 state_dict=state_dict,
                 epoch=context.epoch,
-                iteration=context.iteration,
-                extra=extra,
+                consumed_samples=context.consumed_samples,
                 config=context.model_config,
+                extra=extra,
+                meta=meta,
             )
             context.checkpoint.save(save_path)
 
     def on_batch_end(self, context: TrainContext):
-        if context.iteration - self.last_ckpt_iter >= self.interval:
+        if context.optimizer_step - self.last_ckpt_step >= self.interval:
             self._save_checkpoint(context)
 
     def on_train_end(self, context: TrainContext):
-        if context.iteration != self.last_ckpt_iter:
+        if context.optimizer_step != self.last_ckpt_step:
             self._save_checkpoint(context)
 
     def on_error(self, context: TrainContext):
@@ -201,20 +199,24 @@ class ProgressBarCallback(TrainCallback):
 
     @only_on_rank(0)
     def on_epoch_begin(self, context: TrainContext):
+        total_steps = len(context.dataloader) // context.executor.grad_accum_steps
         self.progress_bar = tqdm(
-            context.dataloader,
+            total=total_steps,
             desc=f"Epoch {context.epoch + 1}/{self.num_epoch}",
             dynamic_ncols=True,
             file=self.file or sys.stdout,
         )
 
     @only_on_rank(0)
-    def on_batch_end(self, context: TrainContext):
+    def on_optimizer_step(self, context: TrainContext):
         postfix = {
+            "step": context.optimizer_step,
             "loss": f"{context.loss:.4f}",
             "lr": f"{context.optimizer.param_groups[-1]['lr']:.2e}",
         }
-        if context.val_loss > 0:
+        if context.grad_norm is not None:
+            postfix["grad_norm"] = f"{context.grad_norm:.2f}"
+        if context.val_loss is not None:
             postfix["val_loss"] = f"{context.val_loss:.4f}"
         self.progress_bar.set_postfix(postfix)
         self.progress_bar.update(1)
@@ -226,19 +228,20 @@ class ProgressBarCallback(TrainCallback):
             self.progress_bar.close()
 
 
-@CallbackFactory.register("metric_logger")
-class MetricLoggerCallback(TrainCallback):
+@CallbackFactory.register("metric")
+class MetricCallback(TrainCallback):
     def __init__(
         self,
         log_dir: str,
         save_interval: int,
-        log_interval: int = 10,
         metrics: List[str] = None,
+        val_step: int = 0,
     ):
-        self.last_log_iter = 0
+        self.last_log_flush_step = 0
         self.save_interval = save_interval
-        self.log_interval = log_interval
         self.metrics = metrics or ["loss", "lr"]
+        self.val_step = val_step
+        self._next_val_step = 0
 
         self.log_dir = Path(log_dir) if log_dir else Path.cwd() / "logs"
         self.log_dir.mkdir(parents=True, exist_ok=True)
@@ -250,53 +253,28 @@ class MetricLoggerCallback(TrainCallback):
             "lr": ctx_get_lr,
             "val_loss": ctx_get_val_loss,
             "grad_norm": ctx_get_grad_norm,
-            "grad_std": ctx_get_grad_std,
-            "grad_max": ctx_get_grad_max,
-            "grad_min": ctx_get_grad_min,
-            "grad_mean": ctx_get_grad_mean,
-            "grad_nan_num": ctx_get_grad_nan_num,
         }
 
-    def _get_log_data(self, context: TrainContext):
+    def _metrics(self, context: TrainContext, names):
         return {
+            m: self._metric_funcs[m](context)
+            for m in names
+            if self._metric_funcs[m](context) is not None
+        }
+
+    @only_on_rank(0)
+    def _append(self, event_type: str, context: TrainContext, **extra):
+        entry = {
+            "type": event_type,
             "timestamp": time.strftime("%Y-%m-%dT%H:%M:%S"),
             "epoch": context.epoch,
-            "iter": context.iteration,
-            **{m: self._metric_funcs[m](context) for m in self.metrics},
+            "step": context.optimizer_step,
+            "consumed_samples": context.consumed_samples,
+            **extra,
         }
+        self.log_cache.append(entry)
 
-    @only_on_rank(0)
-    def _add_log(self, log_data):
-        self.log_cache.append(log_data)
-
-    @only_on_rank(0)
-    def _save_log(self, epoch, iter):
-        log_file = self.log_dir / f"epoch_{epoch}_iter_{iter}_metric.jsonl"
-
-        with open(log_file, "w") as f:
-            for log in self.log_cache:
-                f.write(json.dumps(log) + "\n")
-
-    def on_batch_end(self, context):
-        if context.iteration % self.log_interval == 0:
-            log_data = self._get_log_data(context)
-            self._add_log(log_data)
-
-        if context.iteration - self.last_log_iter >= self.save_interval:
-            self._save_log(context.epoch, context.iteration)
-            self.last_log_iter = context.iteration
-
-    def on_train_end(self, context):
-        if context.iteration != self.last_log_iter:
-            self._save_log(context.epoch, context.iteration)
-
-    def on_error(self, context):
-        self._save_log(context.epoch, context.iteration)
-
-
-@CallbackFactory.register("validation")
-class ValidationCallback(TrainCallback):
-    def _run_validation(self, context: TrainContext):
+    def _run_validation(self, context: TrainContext) -> float:
         context.model.eval()
 
         total_loss = 0.0
@@ -308,27 +286,49 @@ class ValidationCallback(TrainCallback):
                 total_loss += loss.item()
                 num_batches += 1
 
-        avg_loss = total_loss / max(num_batches, 1)
-
         if context.world_size > 1 and dist.is_initialized():
-            loss_tensor = torch.tensor([avg_loss], device=get_current_device())
-            dist.all_reduce(loss_tensor, op=dist.ReduceOp.AVG)
-            avg_loss = loss_tensor.item()
+            stats = torch.tensor(
+                [total_loss, float(num_batches)], device=get_current_device()
+            )
+            dist.all_reduce(stats, op=dist.ReduceOp.SUM)
+            avg_loss = (stats[0] / stats[1]).item()
+        else:
+            avg_loss = total_loss / max(num_batches, 1)
 
-        context.val_loss = avg_loss
         context.model.train()
+        return avg_loss
 
-        step_count = context.iteration // context.config.grad_accum_steps
-        logger.info(
-            f"Epoch {context.epoch + 1}, Step {step_count}, Val Loss: {avg_loss:.4f}"
-        )
+    @only_on_rank(0)
+    def _flush(self, epoch, step):
+        log_file = self.log_dir / f"epoch_{epoch}_step_{step}_metric.jsonl"
+        log_file.parent.mkdir(parents=True, exist_ok=True)
+        with open(log_file, "w") as f:
+            for log in self.log_cache:
+                f.write(json.dumps(log) + "\n")
 
-    def on_optimizer_step(self, context: TrainContext):
-        if context.val_dataloader is None:
-            return
-        cfg = context.config
-        if cfg.val_step <= 0:
-            return
-        step_count = context.iteration // cfg.grad_accum_steps
-        if step_count % cfg.val_step == 0:
-            self._run_validation(context)
+    def on_optimizer_step(self, context):
+        if (
+            context.val_dataloader is not None
+            and self.val_step > 0
+            and context.optimizer_step >= self._next_val_step
+        ):
+            context.val_loss = self._run_validation(context)
+            self._next_val_step = context.optimizer_step + self.val_step
+            self._append("validation", context, val_loss=context.val_loss)
+
+        step_metrics = [m for m in self.metrics if m != "val_loss"]
+        self._append("step", context, **self._metrics(context, step_metrics))
+
+        if context.optimizer_step - self.last_log_flush_step >= self.save_interval:
+            self._flush(context.epoch, context.optimizer_step)
+            self.last_log_flush_step = context.optimizer_step
+
+    def on_epoch_end(self, context):
+        self._append("epoch", context)
+
+    def on_train_end(self, context):
+        if context.optimizer_step != self.last_log_flush_step:
+            self._flush(context.epoch, context.optimizer_step)
+
+    def on_error(self, context):
+        self._flush(context.epoch, context.optimizer_step)

astrai/trainer/train_context.py

@@ -1,9 +1,10 @@
 from dataclasses import dataclass, field
 from pathlib import Path
-from typing import Optional, Self
+from typing import Any, Dict, Optional, Self
 
+import torch
 import torch.nn as nn
-from torch.utils.data import DataLoader
+from torch.utils.data import DataLoader, random_split
 
 from astrai.config.train_config import TrainConfig
 from astrai.dataset import ResumableDistributedSampler
@@ -11,7 +12,7 @@ from astrai.model.components.lora import inject_lora
 from astrai.parallel.executor import BaseExecutor, ExecutorFactory
 from astrai.parallel.setup import get_current_device, get_rank, get_world_size
 from astrai.protocols import OptimizerProtocol, SchedulerProtocol
-from astrai.serialization import Checkpoint, load_json, load_model_weights
+from astrai.serialization import Checkpoint, load_json
 from astrai.trainer.strategy import BaseStrategy, StrategyFactory
 
 
@@ -28,14 +29,23 @@ class TrainContext:
     executor: BaseExecutor = field(default=None)
 
     epoch: int = field(default=0)
-    iteration: int = field(default=0)
+    consumed_samples: int = field(default=0)
     loss: float = field(default=0.0)
-    val_dataloader: DataLoader = field(default=None)
-    val_loss: float = field(default=0.0)
+    grad_norm: Optional[float] = field(default=None)
+    val_dataloader: Optional[DataLoader] = field(default=None)
+    val_loss: Optional[float] = field(default=None)
 
     world_size: int = field(default=1)
     rank: int = field(default=0)
-    kwargs: dict = field(default_factory=dict)
+    kwargs: Dict[str, Any] = field(default_factory=dict)
+
+    @property
+    def optimizer_step(self) -> int:
+        return self.consumed_samples // (
+            self.config.batch_per_device
+            * self.world_size
+            * self.config.grad_accum_steps
+        )
 
 
 class TrainContextBuilder:
@@ -81,21 +91,18 @@ class TrainContextBuilder:
             executor=executor,
         )
 
-        if self._resume_dir is not None:
-            resume_path = Path(self._resume_dir)
-            if (resume_path / "meta.json").exists():
-                checkpoint = Checkpoint.load(self._resume_dir)
-                state_dict = checkpoint.state_dict
+        if self._resume_dir:
+            checkpoint = Checkpoint.load_any(self._resume_dir)
+            if checkpoint is not None:
+                model.load_state_dict(checkpoint.state_dict, strict=False)
                 if checkpoint.config:
                     context.model_config = checkpoint.config
-            else:
-                checkpoint = None
-                state_dict = load_model_weights(self._resume_dir)
-            model.load_state_dict(state_dict, strict=False)
-            if checkpoint is not None:
-                context.epoch = cfg.start_epoch
-                context.iteration = cfg.start_batch
-            context.checkpoint = checkpoint
+                context.epoch = checkpoint.epoch or cfg.start_epoch
+                if checkpoint.consumed_samples > 0:
+                    context.consumed_samples = checkpoint.consumed_samples
+                else:
+                    context.consumed_samples = cfg.start_samples * context.world_size
+                context.checkpoint = checkpoint
 
         if cfg.lora is not None:
             inject_lora(
@@ -108,15 +115,27 @@ class TrainContextBuilder:
         context.optimizer = cfg.optimizer_fn(model)
         context.scheduler = cfg.scheduler_fn(context.optimizer)
 
-        sampler_offset = context.iteration * cfg.batch_per_device
+        train_dataset = cfg.dataset
+        val_dataset = cfg.val_dataset
+
+        if val_dataset is None and cfg.val_split is not None:
+            n_total = len(cfg.dataset)
+            n_val = max(1, int(n_total * cfg.val_split))
+            n_train = n_total - n_val
+            generator = torch.Generator().manual_seed(cfg.random_seed)
+            train_dataset, val_dataset = random_split(
+                cfg.dataset, [n_train, n_val], generator=generator
+            )
+
+        sampler_offset = context.consumed_samples // context.world_size
         sampler = ResumableDistributedSampler(
-            data_source=cfg.dataset,
+            data_source=train_dataset,
             start_epoch=context.epoch,
             start_iter=sampler_offset,
             seed=cfg.random_seed,
         )
         context.dataloader = DataLoader(
-            cfg.dataset,
+            train_dataset,
             batch_size=cfg.batch_per_device,
             sampler=sampler,
             num_workers=cfg.num_workers,
@@ -124,16 +143,16 @@ class TrainContextBuilder:
             prefetch_factor=cfg.prefetch_factor,
         )
 
-        if cfg.val_dataset is not None:
+        if val_dataset is not None:
             val_sampler = ResumableDistributedSampler(
-                data_source=cfg.val_dataset,
+                data_source=val_dataset,
                 start_epoch=0,
                 start_iter=0,
                 seed=cfg.random_seed,
                 shuffle=False,
             )
             context.val_dataloader = DataLoader(
-                cfg.val_dataset,
+                val_dataset,
                 batch_size=cfg.batch_per_device,
                 sampler=val_sampler,
                 num_workers=cfg.num_workers,
@@ -159,9 +178,11 @@ class TrainContextBuilder:
                         obj.load_state_dict(extra[name])
 
         context.strategy = StrategyFactory.create(
+            cfg.strategy,
             model=context.model,
-            train_type=cfg.strategy,
             device=device,
+            executor=executor,
+            model_fn=cfg.model_fn,
             **cfg.extra_kwargs,
         )
 

astrai/trainer/trainer.py

@@ -35,15 +35,14 @@ class Trainer:
                 cfg.ckpt_interval,
             ),
             CallbackFactory.create(
-                "metric_logger",
+                "metric",
                 log_dir=cfg.log_dir,
                 save_interval=cfg.ckpt_interval,
-                log_interval=cfg.log_interval,
                 metrics=cfg.metrics,
+                val_step=cfg.val_step,
             ),
             CallbackFactory.create("progress_bar", cfg.n_epoch),
             CallbackFactory.create("gradient_clipping", cfg.max_grad_norm),
-            CallbackFactory.create("validation"),
         ]
         return callbacks
 
@@ -68,14 +67,15 @@ class Trainer:
                 self._call_callbacks("on_epoch_begin", context)
 
                 for batch in context.dataloader:
-                    self._call_callbacks("on_batch_begin", context)
-
                     with executor.accumulate(context.model):
+                        self._call_callbacks("on_batch_begin", context)
                         loss = context.strategy(batch)
                         context.loss = loss.item()
                         stand_loss = loss / executor.grad_accum_steps
                         executor.backward(stand_loss)
-                        context.iteration += 1
+                        context.consumed_samples += (
+                            context.config.batch_per_device * context.world_size
+                        )
                         self._call_callbacks("on_batch_end", context)
 
                         if executor.sync_gradients:

csrc/__init__.py

@@ -0,0 +1,2 @@
+# Source directory for CUDA kernels — build-time only.
+# Compiled .so files live in astrAI/_ext/.

csrc/build.py

@@ -0,0 +1,48 @@
+from pathlib import Path
+
+
+def _arch_flags() -> list[str]:
+    import torch
+
+    if torch.cuda.is_available():
+        cap = torch.cuda.get_device_capability()
+    else:
+        cap = (8, 0)
+    ver = f"{cap[0]}{cap[1]}"
+    flags = [f"-gencode=arch=compute_{ver},code=sm_{ver}"]
+    # tensor-core mma path (mma.sync.m16n8k16.bf16) requires sm_80+; decide the
+    # kernel dispatch at build time via this define rather than at runtime.
+    if cap[0] < 8:
+        flags.append("-DASTRAI_NO_MMA")
+    return flags
+
+
+_kernels_dir = Path("csrc/kernels")
+REGISTRY: dict[str, dict] = {}
+
+CXX_FLAGS = ["-O3", "-march=native", "-funroll-loops"]
+NVCC_FLAGS = [
+    "-O3",
+    "--expt-relaxed-constexpr",
+    "--use_fast_math",
+    "--ptxas-options=-O3,-v",
+    "--extra-device-vectorization",
+    "--threads=8",
+]
+
+
+def register(name: str, sources: list[str] | None = None, **kwargs):
+    if sources is None:
+        sources = [str(_kernels_dir / f"{name}.cu")]
+    REGISTRY[name] = {
+        "sources": sources,
+        "cxx_flags": [*CXX_FLAGS],
+        "nvcc_flags": [*NVCC_FLAGS, *_arch_flags()],
+        "extra_link_args": kwargs.pop("extra_link_args", []),
+        **kwargs,
+    }
+
+
+register("attn_decode")
+register("attn_prefill")
+register("attn_paged_decode")

csrc/kernels/attn_common.h

@@ -0,0 +1,54 @@
+#pragma once
+
+
+template<typename T, typename AT = float>
+struct AttentionParams {
+    int batch;
+    int q_head;
+    int kv_head;
+    int q_len;
+    int kv_len;
+    int head_dim;
+    int use_mask;
+    int is_causal;
+    int causal_offset;
+    int num_splits;
+    float scale;
+
+    const T* __restrict__ q;
+    const T* __restrict__ k;
+    const T* __restrict__ v;
+    const bool* __restrict__ mask;
+
+    T* __restrict__ o;
+    AT* __restrict__ o_part;
+    AT* __restrict__ ml_part;
+};
+
+template<typename T, typename AT = float>
+struct PagedAttentionParams {
+    int batch;
+    int q_head;
+    int kv_head;
+    int q_len;
+    int kv_len;
+    int head_dim;
+    int use_mask;
+    int is_causal;
+    int causal_offset;
+    float scale;
+
+    int num_splits;
+    int page_size;
+    int max_pages;
+
+    const T* __restrict__ q;
+    const T* __restrict__ k_cache;
+    const T* __restrict__ v_cache;
+    const bool* __restrict__ mask;
+    const int64_t* __restrict__ page_table;
+    
+    T* __restrict__ o;
+    AT* __restrict__ o_part;
+    AT* __restrict__ ml_part;
+};

csrc/kernels/attn_decode.cu

@@ -0,0 +1,111 @@
+#include "attn_decode_split_kv.cuh"
+#include "attn_entry_utils.cuh"
+
+#ifndef ASTRAI_NO_MMA
+#include "attn_decode_split_kv_mma.cuh"
+#endif
+
+static int decode_num_splits(int base_blocks, int tiles_total) {
+    int sm_count = 0;
+    cudaDeviceGetAttribute(&sm_count, cudaDevAttrMultiProcessorCount, 0);
+    int n = (2 * sm_count + base_blocks - 1) / base_blocks;
+    return std::max(1, std::min(n, std::min(tiles_total, 32)));
+}
+
+// Scalar fallback: one warp per query head, split-KV across grid.z.
+static void launch_scalar_decode(AttentionParams<bf16>& p) {
+    int group_size = p.q_head / p.kv_head;
+    int chunks_total = (p.kv_len + DC_CHUNK - 1) / DC_CHUNK;
+    p.num_splits = decode_num_splits(p.batch * p.kv_head, chunks_total);
+
+    auto fopt = torch::TensorOptions().dtype(torch::kFloat32).device(torch::kCUDA);
+    auto o_part = torch::empty({p.batch, p.q_head, p.num_splits, p.head_dim}, fopt);
+    auto ml_part = torch::empty({p.batch, p.q_head, p.num_splits, 2}, fopt);
+    p.o_part = o_part.data_ptr<float>();
+    p.ml_part = ml_part.data_ptr<float>();
+
+    size_t smem = DC_CHUNK * p.head_dim * sizeof(bf16);
+    attn_decode_split_kv_kernel<<<dim3(p.batch * p.kv_head, 1, p.num_splits), dim3(32, group_size), smem>>>(p);
+    attn_decode_combine_kernel<<<p.batch * p.q_head, p.head_dim>>>(p);
+}
+
+#ifndef ASTRAI_NO_MMA
+// MMA head-packing requires G <= 16 (sQ has BR=16 rows). sm_80+ tensor-core
+// + cp.async wins even at G=1 (decode is memory-bound, not compute-bound).
+template <int HEAD_DIM, int BC>
+static void launch_mma_decode(AttentionParams<bf16>& p) {
+    int tiles_total = (p.kv_len + BC - 1) / BC;
+    p.num_splits = decode_num_splits(p.batch * p.kv_head, tiles_total);
+
+    auto fopt = torch::TensorOptions().dtype(torch::kFloat32).device(torch::kCUDA);
+    auto o_part = torch::empty({p.batch, p.q_head, p.num_splits, p.head_dim}, fopt);
+    auto ml_part = torch::empty({p.batch, p.q_head, p.num_splits, 2}, fopt);
+    p.o_part = o_part.data_ptr<float>();
+    p.ml_part = ml_part.data_ptr<float>();
+
+    attn_decode_split_kv_mma_kernel<HEAD_DIM, BC>
+        <<<dim3(p.kv_head, p.batch, p.num_splits), 32>>>(p);
+    attn_decode_combine_kernel<<<p.batch * p.q_head, p.head_dim>>>(p);
+}
+#endif
+
+template <int HEAD_DIM>
+static void dispatch_decode(AttentionParams<bf16>& p) {
+#ifndef ASTRAI_NO_MMA
+    int G = p.q_head / p.kv_head;
+    if (!p.use_mask && G >= 1 && G <= 16) {
+        launch_mma_decode<HEAD_DIM, 32>(p);
+        return;
+    }
+#endif
+    launch_scalar_decode(p);
+}
+
+torch::Tensor attn_decode(
+    torch::Tensor q,
+    torch::Tensor k,
+    torch::Tensor v,
+    c10::optional<torch::Tensor> mask,
+    bool is_causal = false,
+    int64_t causal_offset = 0,
+    c10::optional<double> scale = c10::nullopt
+) {
+    AttentionParams<bf16> p;
+    attn_pack_params(q, k, v, mask, is_causal, causal_offset, scale, p);
+    TORCH_CHECK(p.q_len == 1, "Q seq_len must be 1");
+    TORCH_CHECK(p.head_dim % 32 == 0, "head_dim must be multiple of 32");
+
+    auto O = torch::empty_like(q);
+    p.o = (bf16*)O.data_ptr();
+
+    switch (p.head_dim) {
+        case 32:
+            dispatch_decode<32>(p);
+            break;
+        case 64:
+            dispatch_decode<64>(p);
+            break;
+        case 128:
+            dispatch_decode<128>(p);
+            break;
+        case 256:
+            dispatch_decode<256>(p);
+            break;
+        default:
+            TORCH_CHECK(false, "decode: unsupported head_dim ", p.head_dim,
+                        " (supported: 32, 64, 128, 256)");
+    }
+    return O;
+}
+
+PYBIND11_MODULE(TORCH_EXTENSION_NAME, m) {
+    m.def("attn_decode", &attn_decode,
+        py::arg("q"),
+        py::arg("k"),
+        py::arg("v"),
+        py::arg("mask") = py::none(),
+        py::arg("is_causal") = false,
+        py::arg("causal_offset") = 0,
+        py::arg("scale") = py::none(),
+        "GQA decode (tensor-core head-packing on sm_80+, scalar fallback)");
+}

csrc/kernels/attn_decode_split_kv.cuh

@@ -0,0 +1,116 @@
+#pragma once
+#include <cuda_bf16.h>
+#include <float.h>
+#include "attn_common.h"
+
+using bf16 = __nv_bfloat16;
+constexpr int DC_CHUNK = 64;
+
+__device__ inline float warp_reduce_sum(float val) {
+    for (int offset = 16; offset > 0; offset >>= 1)
+        val += __shfl_xor_sync(0xFFFFFFFF, val, offset);
+    return val;
+}
+
+__global__ void attn_decode_split_kv_kernel(AttentionParams<bf16> p) {
+    int batch = blockIdx.x / p.kv_head;
+    int kv_head = blockIdx.x % p.kv_head;
+    int split = blockIdx.z;
+    int group_size = blockDim.y;
+    int q_head = kv_head * group_size + threadIdx.y;
+    int lane = threadIdx.x;
+    int hd_per_thread = p.head_dim / 32;
+
+    float q_reg[8];
+    int q_off = ((batch * p.q_head + q_head) * 1) * p.head_dim + lane * hd_per_thread;
+    for (int i = 0; i < hd_per_thread; i++)
+        q_reg[i] = __bfloat162float(p.q[q_off + i]);
+
+    int kv_base = ((batch * p.kv_head + kv_head) * p.kv_len) * p.head_dim;
+    int mask_base = batch * p.kv_len;
+
+    float m = -FLT_MAX, d = 0.0f, acc_reg[8] = {0.0f};
+
+    extern __shared__ __align__(16) bf16 k_smem[];
+
+    // Split-KV: each split processes a contiguous subset of chunks
+    int chunks_total = (p.kv_len + DC_CHUNK - 1) / DC_CHUNK;
+    int chunks_per_split = (chunks_total + p.num_splits - 1) / p.num_splits;
+    int ch_begin = split * chunks_per_split;
+    int ch_end = min(chunks_total, ch_begin + chunks_per_split);
+
+    for (int ci = ch_begin; ci < ch_end; ci++) {
+        int chunk_start = ci * DC_CHUNK;
+        int this_chunk = min(DC_CHUNK, p.kv_len - chunk_start);
+
+        int total = this_chunk * p.head_dim;
+        for (int i = threadIdx.y * 32 + lane; i < total; i += blockDim.x * blockDim.y)
+            k_smem[i] = p.k[kv_base + chunk_start * p.head_dim + i];
+        __syncthreads();
+
+        for (int s = 0; s < this_chunk; s++) {
+            float partial = 0.0f;
+            for (int i = 0; i < hd_per_thread; i++)
+                partial += q_reg[i] * __bfloat162float(k_smem[s * p.head_dim + lane * hd_per_thread + i]);
+            partial = warp_reduce_sum(partial) * p.scale;
+
+            if (p.use_mask && p.mask && !p.mask[mask_base + chunk_start + s])
+                partial = -FLT_MAX;
+            if (p.is_causal && (chunk_start + s) > p.causal_offset)
+                partial = -FLT_MAX;
+
+            float new_m = fmaxf(m, partial);
+            float alpha = expf(m - new_m);
+            float beta  = expf(partial - new_m);
+            d = d * alpha + beta;
+
+            int v_off = kv_base + (chunk_start + s) * p.head_dim + lane * hd_per_thread;
+            for (int i = 0; i < hd_per_thread; i++)
+                acc_reg[i] = acc_reg[i] * alpha + __bfloat162float(p.v[v_off + i]) * beta;
+            m = new_m;
+        }
+        __syncthreads();
+    }
+
+    // ---- write UN-normalised partials for this split ----
+    size_t bh = (size_t)batch * p.q_head + q_head;
+    size_t slot = bh * p.num_splits + split;
+    int d0 = lane * hd_per_thread;
+    for (int i = 0; i < hd_per_thread; i++) {
+        int dd = d0 + i;
+        p.o_part[slot * p.head_dim + dd] = acc_reg[i];
+    }
+    if (lane == 0) {
+        p.ml_part[slot * 2] = m;
+        p.ml_part[slot * 2 + 1] = d;
+    }
+}
+
+// Reduce split-K partials into the final bf16 output. One block per (batch,
+// q_head); each thread folds across all splits with a single-pass
+// online-rescale reduction (expf + FMA counts halved vs 3-pass original).
+__global__ void attn_decode_combine_kernel(AttentionParams<bf16> p) {
+    int bh = blockIdx.x;
+    int d = threadIdx.x;
+    if (d >= p.head_dim) return;
+
+    size_t split_base = (size_t)bh * p.num_splits;
+    const float* mlp = p.ml_part + split_base * 2;
+    const float* op = p.o_part + split_base * p.head_dim;
+
+    float m = -FLT_MAX, l = 0.0f, acc = 0.0f;
+    for (int s = 0; s < p.num_splits; s++) {
+        float mi = mlp[s * 2];
+        if (mi <= -FLT_MAX) continue;
+        float li = mlp[s * 2 + 1];
+        float nm = fmaxf(m, mi);
+        float corr = __expf(m - nm);
+        float e = __expf(mi - nm);
+        acc = acc * corr + op[s * p.head_dim + d] * e;
+        l = l * corr + li * e;
+        m = nm;
+    }
+
+    float inv = (l > 1e-20f) ? (1.0f / l) : 0.0f;
+    p.o[(size_t)bh * p.head_dim + d] = __float2bfloat16(acc * inv);
+}

csrc/kernels/attn_decode_split_kv_mma.cuh

@@ -0,0 +1,172 @@
+#pragma once
+#include <cfloat>
+#include <cuda_bf16.h>
+#include "attn_common.h"
+#include "attn_mma_utils.cuh"
+
+using bf16 = __nv_bfloat16;
+
+// Split-K (FlashDecoding) tensor-core decode via GQA head-packing.
+//
+// Decode has q_len == 1, so S = q @ K^T is a GEMV per head — no tensor-core work
+// on its own. But GQA gives us G = q_head / kv_head query heads that all share
+// one kv_head. We pack those G heads into the M=16 rows of mma.sync.m16n8k16,
+// turning G independent GEMVs into a single GEMM that reuses each loaded K/V tile
+// across all G heads (K/V load is the decode bottleneck, so the reuse is the win,
+// not the flops). The KV sequence is partitioned across gridDim.z blocks so that
+// a decode with only batch*kv_head independent tasks can fill all SMs. Each
+// (batch, kv_head, split) block computes an UN-normalised partial (Oacc, m, l)
+// over its KV slice; the combine kernel below reduces across splits. Fixes the
+// "grid too small" bottleneck (0.04 waves/SM → many blocks) for long-context,
+// small-batch decode.
+//
+// Partial layout (float, contiguous):
+//   o_part : [batch, q_head, num_splits, HEAD_DIM]
+//   ml_part: [batch, q_head, num_splits, 2]  (m, l)
+//
+// Optimizations:
+//   - cp.async global→shared for K/V (bypasses registers, cuts instruction count)
+//   - XOR swizzle (swiz_col): LD=HEAD_DIM, zero waste, no bank conflicts
+//   - pre-scaled Q: Q scaled during load, softmax skips per-tile multiply
+//   - single-buffer: keeps smem small for high occupancy
+template <int HEAD_DIM, int BC>
+__global__ void attn_decode_split_kv_mma_kernel(AttentionParams<bf16> p) {
+    constexpr int BR = 16;
+    constexpr int KD = HEAD_DIM / 16;
+    constexpr int NC8 = BC / 8;
+    constexpr int KT2 = BC / 16;
+    constexpr int DN8 = HEAD_DIM / 8;
+    constexpr int LD = HEAD_DIM;
+    constexpr int SWIZ_MASK = (HEAD_DIM >= 64) ? 7 : (HEAD_DIM / 8 - 1);
+
+    const int lane = threadIdx.x;
+    const int gid = lane >> 2;
+    const int tid4 = lane & 3;
+
+    const int kv_head = blockIdx.x;
+    const int batch = blockIdx.y;
+    const int split = blockIdx.z;
+    const int G = p.q_head / p.kv_head;
+    const int q_head0 = kv_head * G;
+
+    __shared__ __align__(16) bf16 sK[BC * HEAD_DIM];
+    __shared__ __align__(16) bf16 sV[BC * HEAD_DIM];
+    __shared__ __align__(16) bf16 sQ[BR * HEAD_DIM];
+
+    for (int i = lane; i < BR * HEAD_DIM; i += 32) {
+        int r = i / HEAD_DIM, d = i % HEAD_DIM;
+        bf16 val = __float2bfloat16(0.0f);
+        if (r < G) {
+            int qh = q_head0 + r;
+            val = p.q[(batch * p.q_head + qh) * HEAD_DIM + d];
+        }
+        sQ[r * LD + swiz_col(d, r, SWIZ_MASK)] = val;
+    }
+    __syncwarp();
+
+    unsigned Qa[KD][4];
+    int qrow_l = (lane & 7) + (lane & 8);
+    int qcol_l = (lane & 16) ? 8 : 0;
+#pragma unroll
+    for (int kt = 0; kt < KD; kt++)
+        ldmatrix_x4(Qa[kt], &sQ[qrow_l * LD + swiz_col(kt * 16 + qcol_l, qrow_l, SWIZ_MASK)]);
+
+    float Oacc[DN8][4];
+#pragma unroll
+    for (int j = 0; j < DN8; j++)
+        Oacc[j][0] = Oacc[j][1] = Oacc[j][2] = Oacc[j][3] = 0.0f;
+    float m0 = -FLT_MAX, m1 = -FLT_MAX, l0 = 0.0f, l1 = 0.0f;
+
+    const int kv_base = (batch * p.kv_head + kv_head) * p.kv_len * HEAD_DIM;
+    const int mask_base = batch * p.kv_len;
+    const int tiles_total = (p.kv_len + BC - 1) / BC;
+    const int tiles_per_split = (tiles_total + p.num_splits - 1) / p.num_splits;
+    const int ti_begin = split * tiles_per_split;
+    const int ti_end = min(tiles_total, ti_begin + tiles_per_split);
+    const int has_mask = p.use_mask && p.mask;
+
+    for (int ti = ti_begin; ti < ti_end; ti++) {
+        int kv0 = ti * BC;
+
+        bool full_tile = (kv0 + BC <= p.kv_len);
+        if (full_tile) {
+            constexpr int VEC = 8;
+            int total = BC * HEAD_DIM;
+#pragma unroll
+            for (int i = lane * VEC; i < total; i += 32 * VEC) {
+                int r = i / HEAD_DIM, d = i % HEAD_DIM;
+                int kc = kv0 + r;
+                cp_async_16(&sK[r * LD + swiz_col(d, r, SWIZ_MASK)],
+                            &p.k[kv_base + kc * HEAD_DIM + d]);
+                cp_async_16(&sV[r * LD + swiz_col(d, r, SWIZ_MASK)],
+                            &p.v[kv_base + kc * HEAD_DIM + d]);
+            }
+            cp_async_commit();
+            cp_async_wait_all();
+        } else {
+            for (int i = lane; i < BC * HEAD_DIM; i += 32) {
+                int r = i / HEAD_DIM, d = i % HEAD_DIM;
+                int kc = kv0 + r;
+                bf16 z = __float2bfloat16(0.0f);
+                sK[r * LD + swiz_col(d, r, SWIZ_MASK)] =
+                    (kc < p.kv_len) ? p.k[kv_base + kc * HEAD_DIM + d] : z;
+                sV[r * LD + swiz_col(d, r, SWIZ_MASK)] =
+                    (kc < p.kv_len) ? p.v[kv_base + kc * HEAD_DIM + d] : z;
+            }
+        }
+        __syncwarp();
+
+        float Sacc[NC8][4];
+        mma_compute_scores<KD, NC8>(Qa, sK, LD, SWIZ_MASK, lane, Sacc);
+
+        #pragma unroll
+        for (int n8 = 0; n8 < NC8; n8++)
+            Sacc[n8][0] *= p.scale, Sacc[n8][1] *= p.scale,
+            Sacc[n8][2] *= p.scale, Sacc[n8][3] *= p.scale;
+
+        int maxc = p.is_causal ? min(p.kv_len, p.causal_offset + 1) : p.kv_len;
+        mma_softmax_tile<NC8, DN8>(kv0, maxc, maxc,
+                                    mask_base, p.mask, has_mask,
+                                    Sacc, Oacc, m0, m1, l0, l1, lane);
+
+        mma_pv_accumulate<DN8, KT2>(Sacc, sV, LD, SWIZ_MASK, lane, Oacc);
+        __syncwarp();
+    }
+
+    // ---- write UN-normalised partials for this split ----
+    auto split_slot = [&](int h) -> size_t {
+        size_t bh = (size_t)batch * p.q_head + h;
+        return bh * p.num_splits + split;
+    };
+#pragma unroll
+    for (int dn8 = 0; dn8 < DN8; dn8++) {
+        int d = dn8 * 8 + 2 * tid4;
+        int r0 = gid, r1 = gid + 8;
+        if (r0 < G) {
+            int h = q_head0 + r0;
+            float* op = p.o_part + split_slot(h) * HEAD_DIM;
+            op[d] = Oacc[dn8][0];
+            op[d + 1] = Oacc[dn8][1];
+        }
+        if (r1 < G) {
+            int h = q_head0 + r1;
+            float* op = p.o_part + split_slot(h) * HEAD_DIM;
+            op[d] = Oacc[dn8][2];
+            op[d + 1] = Oacc[dn8][3];
+        }
+    }
+    if (tid4 == 0) {
+        int r0 = gid, r1 = gid + 8;
+        if (r0 < G) {
+            int h = q_head0 + r0;
+            float* mp = p.ml_part + split_slot(h) * 2;
+            mp[0] = m0; mp[1] = l0;
+        }
+        if (r1 < G) {
+            int h = q_head0 + r1;
+            float* mp = p.ml_part + split_slot(h) * 2;
+            mp[0] = m1; mp[1] = l1;
+        }
+    }
+}
+

csrc/kernels/attn_entry_utils.cuh

@@ -0,0 +1,43 @@
+#pragma once
+#include <torch/extension.h>
+#include "attn_common.h"
+
+template<typename T>
+inline void attn_pack_params(
+    torch::Tensor q,
+    torch::Tensor k,
+    torch::Tensor v,
+    c10::optional<torch::Tensor> mask,
+    bool is_causal,
+    int64_t causal_offset,
+    c10::optional<double> scale,
+    AttentionParams<T>& p
+) {
+    TORCH_CHECK(q.is_cuda() && k.is_cuda() && v.is_cuda());
+    TORCH_CHECK(q.dtype() == torch::kBFloat16);
+    TORCH_CHECK(k.dtype() == torch::kBFloat16);
+    TORCH_CHECK(v.dtype() == torch::kBFloat16);
+
+    p.batch = (int)q.size(0);
+    p.q_head = (int)q.size(1);
+    p.kv_head = (int)k.size(1);
+    p.q_len = (int)q.size(2);
+    p.kv_len = (int)k.size(2);
+    p.head_dim = (int)q.size(3);
+    p.use_mask = mask.has_value() ? 1 : 0;
+    p.is_causal = is_causal ? 1 : 0;
+    p.causal_offset = (int)causal_offset;
+    p.scale = scale.has_value() ? (float)scale.value() : 1.0f / sqrtf((float)p.head_dim);
+    p.q = (const T*)q.data_ptr();
+    p.k = (const T*)k.data_ptr();
+    p.v = (const T*)v.data_ptr();
+    if (p.use_mask) {
+        TORCH_CHECK(mask.value().dtype() == torch::kBool);
+        TORCH_CHECK(mask.value().dim() == 2);
+        TORCH_CHECK(mask.value().size(0) == p.batch);
+        TORCH_CHECK(mask.value().size(1) == p.kv_len);
+        p.mask = mask.value().data_ptr<bool>();
+    } else {
+        p.mask = nullptr;
+    }
+}

csrc/kernels/attn_mma_utils.cuh

@@ -0,0 +1,237 @@
+#pragma once
+#include <cfloat>
+#include <cuda_fp16.h>
+#include <cuda_runtime.h>
+
+// Shared MMA utilities for tensor-core GQA kernels.
+// mma.sync.m16n8k16 PTX wrappers, ldmatrix helpers, and bf16 packing.
+
+// mma.sync.aligned.m16n8k16.row.col.f32.bf16.bf16.f32
+__device__ __forceinline__ void mma16816(float* d, const unsigned* a,
+                                          const unsigned* b, const float* c) {
+    asm volatile(
+        "mma.sync.aligned.m16n8k16.row.col.f32.bf16.bf16.f32 "
+        "{%0,%1,%2,%3}, {%4,%5,%6,%7}, {%8,%9}, {%10,%11,%12,%13};"
+        : "=f"(d[0]), "=f"(d[1]), "=f"(d[2]), "=f"(d[3])
+        : "r"(a[0]), "r"(a[1]), "r"(a[2]), "r"(a[3]), "r"(b[0]), "r"(b[1]),
+          "f"(c[0]), "f"(c[1]), "f"(c[2]), "f"(c[3]));
+}
+
+// read two adjacent bf16 from smem as one packed .b32 (elem0 low, elem1 high)
+__device__ __forceinline__ unsigned ld2(const bf16* p) {
+    return *reinterpret_cast<const unsigned*>(p);
+}
+
+// pack two floats into one bf16x2 as .b32
+__device__ __forceinline__ unsigned pk2(float a, float b) {
+    __nv_bfloat162 v = __floats2bfloat162_rn(a, b);
+    return *reinterpret_cast<unsigned*>(&v);
+}
+
+// pack two (non-contiguous) bf16 into one .b32
+__device__ __forceinline__ unsigned pkb(bf16 a, bf16 b) {
+    __nv_bfloat162 v;
+    v.x = a;
+    v.y = b;
+    return *reinterpret_cast<unsigned*>(&v);
+}
+
+// ldmatrix: cooperatively load mma fragments from smem (one instruction per
+// 16x16 / 16x8 tile) with the exact register layout mma expects — replaces the
+// scalar per-thread fragment packing, cutting shared-load instructions and bank
+// conflicts. Each lane supplies the shared address of one 8-wide row.
+__device__ __forceinline__ void ldmatrix_x4(unsigned* r, const bf16* p) {
+    unsigned a = __cvta_generic_to_shared(p);
+    asm volatile("ldmatrix.sync.aligned.m8n8.x4.shared.b16 {%0,%1,%2,%3}, [%4];"
+                 : "=r"(r[0]), "=r"(r[1]), "=r"(r[2]), "=r"(r[3])
+                 : "r"(a));
+}
+__device__ __forceinline__ void ldmatrix_x2(unsigned* r, const bf16* p) {
+    unsigned a = __cvta_generic_to_shared(p);
+    asm volatile("ldmatrix.sync.aligned.m8n8.x2.shared.b16 {%0,%1}, [%2];"
+                 : "=r"(r[0]), "=r"(r[1])
+                 : "r"(a));
+}
+__device__ __forceinline__ void ldmatrix_x2_trans(unsigned* r, const bf16* p) {
+    unsigned a = __cvta_generic_to_shared(p);
+    asm volatile("ldmatrix.sync.aligned.m8n8.x2.trans.shared.b16 {%0,%1}, [%2];"
+                 : "=r"(r[0]), "=r"(r[1])
+                 : "r"(a));
+}
+
+// XOR swizzle for shared-memory column at 8-bf16 chunk granularity.
+// Eliminates ldmatrix bank conflicts without LD padding: consecutive rows
+// land in distinct bank groups. swiz_col(d, r, mask) = ((d>>3)^(r&mask))<<3 | (d&7).
+// mask must cover log2(HEAD_DIM/8) chunk bits but stay within LD: use 7 for
+// HEAD_DIM>=64 (8+ chunks), 3 for HEAD_DIM=32 (4 chunks). Default 7 keeps
+// existing HEAD_DIM>=64 call sites working unchanged.
+__device__ __forceinline__ int swiz_col(int d, int r, int mask = 7) {
+    return ((d >> 3) ^ (r & mask)) << 3 | (d & 7);
+}
+
+// cp.async: copy 16 bytes (8 bf16) from global to shared memory directly,
+// bypassing registers. Eliminates shared-store bank conflicts and cuts
+// load-loop instruction count in half (1 cp.async vs 1 LDG + 1 STS).
+// Requires sm_80+.
+__device__ __forceinline__ void cp_async_16(bf16* smem_ptr, const void* gmem_ptr) {
+    unsigned smem_addr = __cvta_generic_to_shared(smem_ptr);
+    asm volatile("cp.async.ca.shared.global [%0], [%1], 16;"
+                 :: "r"(smem_addr), "l"(gmem_ptr));
+}
+
+// Predicated cp.async: copy 16 bytes when `pred`, otherwise zero-fill the
+// destination (src-size operand = 0 → no bytes read from src, so an
+// out-of-bounds src address is never dereferenced). Lets full and partial
+// tiles share one uniform async load path — no scalar fallback branch.
+__device__ __forceinline__ void cp_async_16_pred(bf16* smem_ptr,
+                                                  const void* gmem_ptr,
+                                                  bool pred) {
+    unsigned smem_addr = __cvta_generic_to_shared(smem_ptr);
+    int src_size = pred ? 16 : 0;
+    asm volatile("cp.async.ca.shared.global [%0], [%1], 16, %2;"
+                 :: "r"(smem_addr), "l"(gmem_ptr), "r"(src_size));
+}
+
+__device__ __forceinline__ void cp_async_commit() {
+    asm volatile("cp.async.commit_group;");
+}
+
+__device__ __forceinline__ void cp_async_wait_all() {
+    asm volatile("cp.async.wait_all;");
+}
+
+// Wait until at most N commit groups are still in flight. Used for
+// double-buffered pipelining: wait_group<1> lets the next tile's cp.async
+// continue while ensuring the current tile's data is ready.
+template <int N>
+__device__ __forceinline__ void cp_async_wait_group() {
+    asm volatile("cp.async.wait_group %0;" :: "n"(N));
+}
+
+// ---------------------------------------------------------------------------
+// Shared MMA compute functions — used by both decode and prefill MMA kernels.
+// Extracted because S=Q@K^T, online softmax, and P@V are structurally identical
+// between the two kernels; only the per-row causal/mask bounds differ.
+// ---------------------------------------------------------------------------
+
+// S = Q @ K^T  (Qa pre-loaded and pre-scaled by the caller).
+// LD and SWIZ_MASK are constexpr in the calling kernel — passing them as
+// runtime ints lets the compiler fold them while keeping the signature clean.
+template <int KD, int NC8>
+__device__ inline void mma_compute_scores(
+    const unsigned Qa[KD][4],
+    const bf16* __restrict__ sK,
+    int LD, int SWIZ_MASK, int lane,
+    float Sacc[NC8][4])
+{
+    #pragma unroll
+    for (int n8 = 0; n8 < NC8; n8++) {
+        Sacc[n8][0] = Sacc[n8][1] = Sacc[n8][2] = Sacc[n8][3] = 0.0f;
+        int krow_l = n8 * 8 + (lane & 7);
+        int kcol_h = (lane & 8) ? 8 : 0;
+        #pragma unroll
+        for (int kt = 0; kt < KD; kt++) {
+            unsigned b[2];
+            ldmatrix_x2(b, &sK[krow_l * LD + swiz_col(kt * 16 + kcol_h, krow_l, SWIZ_MASK)]);
+            mma16816(Sacc[n8], Qa[kt], b, Sacc[n8]);
+        }
+    }
+}
+
+// Online softmax + Oacc rescale for one K/V tile. maxc0/maxc1 are the per-row
+// KV column bounds — prefill passes per-query-row causal limits while decode
+// passes the same value for both rows (q_len==1). Sacc is consumed in place
+// (replaced by P = exp(S - nm) for the subsequent P@V step).
+template <int NC8, int DN8>
+__device__ inline void mma_softmax_tile(
+    int kv0,
+    int maxc0, int maxc1,
+    int mask_base,
+    const bool* __restrict__ mask,
+    bool has_mask,
+    float Sacc[NC8][4],
+    float Oacc[DN8][4],
+    float& m0, float& m1,
+    float& l0, float& l1,
+    int lane)
+{
+    int tid4 = lane & 3;
+
+    float rmax0 = -FLT_MAX, rmax1 = -FLT_MAX;
+    #pragma unroll
+    for (int n8 = 0; n8 < NC8; n8++) {
+        int cc = kv0 + n8 * 8 + 2 * tid4;
+        int c1 = cc + 1;
+        bool b0 = (cc >= maxc0) || (has_mask && !mask[mask_base + cc]);
+        bool b1 = (c1 >= maxc0) || (has_mask && !mask[mask_base + c1]);
+        bool b2 = (cc >= maxc1) || (has_mask && !mask[mask_base + cc]);
+        bool b3 = (c1 >= maxc1) || (has_mask && !mask[mask_base + c1]);
+        float s0 = b0 ? -FLT_MAX : Sacc[n8][0];
+        float s1 = b1 ? -FLT_MAX : Sacc[n8][1];
+        float s2 = b2 ? -FLT_MAX : Sacc[n8][2];
+        float s3 = b3 ? -FLT_MAX : Sacc[n8][3];
+        Sacc[n8][0] = s0; Sacc[n8][1] = s1;
+        Sacc[n8][2] = s2; Sacc[n8][3] = s3;
+        rmax0 = fmaxf(rmax0, fmaxf(s0, s1));
+        rmax1 = fmaxf(rmax1, fmaxf(s2, s3));
+    }
+    rmax0 = fmaxf(rmax0, __shfl_xor_sync(0xFFFFFFFF, rmax0, 1));
+    rmax0 = fmaxf(rmax0, __shfl_xor_sync(0xFFFFFFFF, rmax0, 2));
+    rmax1 = fmaxf(rmax1, __shfl_xor_sync(0xFFFFFFFF, rmax1, 1));
+    rmax1 = fmaxf(rmax1, __shfl_xor_sync(0xFFFFFFFF, rmax1, 2));
+
+    float nm0 = fmaxf(m0, rmax0), nm1 = fmaxf(m1, rmax1);
+    float corr0 = (nm0 == -FLT_MAX) ? 1.0f : __expf(m0 - nm0);
+    float corr1 = (nm1 == -FLT_MAX) ? 1.0f : __expf(m1 - nm1);
+
+    float rsum0 = 0.0f, rsum1 = 0.0f;
+    #pragma unroll
+    for (int n8 = 0; n8 < NC8; n8++) {
+        float p0 = (Sacc[n8][0] == -FLT_MAX) ? 0.0f : __expf(Sacc[n8][0] - nm0);
+        float p1 = (Sacc[n8][1] == -FLT_MAX) ? 0.0f : __expf(Sacc[n8][1] - nm0);
+        float p2 = (Sacc[n8][2] == -FLT_MAX) ? 0.0f : __expf(Sacc[n8][2] - nm1);
+        float p3 = (Sacc[n8][3] == -FLT_MAX) ? 0.0f : __expf(Sacc[n8][3] - nm1);
+        Sacc[n8][0] = p0; Sacc[n8][1] = p1;
+        Sacc[n8][2] = p2; Sacc[n8][3] = p3;
+        rsum0 += p0 + p1;
+        rsum1 += p2 + p3;
+    }
+    rsum0 += __shfl_xor_sync(0xFFFFFFFF, rsum0, 1);
+    rsum0 += __shfl_xor_sync(0xFFFFFFFF, rsum0, 2);
+    rsum1 += __shfl_xor_sync(0xFFFFFFFF, rsum1, 1);
+    rsum1 += __shfl_xor_sync(0xFFFFFFFF, rsum1, 2);
+    l0 = l0 * corr0 + rsum0;
+    l1 = l1 * corr1 + rsum1;
+    m0 = nm0; m1 = nm1;
+
+    #pragma unroll
+    for (int j = 0; j < DN8; j++) {
+        Oacc[j][0] *= corr0; Oacc[j][1] *= corr0;
+        Oacc[j][2] *= corr1; Oacc[j][3] *= corr1;
+    }
+}
+
+// O += P @ V  (Sacc must contain P = attention weights after softmax).
+template <int DN8, int KT2>
+__device__ inline void mma_pv_accumulate(
+    float Sacc[][4],
+    const bf16* __restrict__ sV,
+    int LD, int SWIZ_MASK, int lane,
+    float Oacc[DN8][4])
+{
+    #pragma unroll
+    for (int kt2 = 0; kt2 < KT2; kt2++) {
+        unsigned Pa[4];
+        Pa[0] = pk2(Sacc[kt2 * 2][0], Sacc[kt2 * 2][1]);
+        Pa[1] = pk2(Sacc[kt2 * 2][2], Sacc[kt2 * 2][3]);
+        Pa[2] = pk2(Sacc[kt2 * 2 + 1][0], Sacc[kt2 * 2 + 1][1]);
+        Pa[3] = pk2(Sacc[kt2 * 2 + 1][2], Sacc[kt2 * 2 + 1][3]);
+        int vrow_l = kt2 * 16 + (lane & 15);
+        #pragma unroll
+        for (int dn8 = 0; dn8 < DN8; dn8++) {
+            unsigned b[2];
+            ldmatrix_x2_trans(b, &sV[vrow_l * LD + swiz_col(dn8 * 8, vrow_l, SWIZ_MASK)]);
+            mma16816(Oacc[dn8], Pa, b, Oacc[dn8]);
+        }
+    }
+}

csrc/kernels/attn_paged_decode.cu

@@ -0,0 +1,152 @@
+#include "attn_paged_decode_split_kv.cuh"
+#ifndef ASTRAI_NO_MMA
+#include "attn_paged_decode_split_kv_mma.cuh"
+#endif
+
+#include <torch/extension.h>
+#include <c10/cuda/CUDAGuard.h>
+
+static int paged_decode_num_splits(int base_blocks, int tiles_total) {
+    int sm_count = 0;
+    cudaDeviceGetAttribute(&sm_count, cudaDevAttrMultiProcessorCount, 0);
+    int n = (2 * sm_count + base_blocks - 1) / base_blocks;
+    return std::max(1, std::min(n, std::min(tiles_total, 32)));
+}
+
+static void launch_paged_scalar_decode(PagedAttentionParams<bf16>& p) {
+    int group_size = p.q_head / p.kv_head;
+    int chunks_total = (p.kv_len + PDC_CHUNK - 1) / PDC_CHUNK;
+    p.num_splits = paged_decode_num_splits(p.batch * p.kv_head, chunks_total);
+
+    auto fopt = torch::TensorOptions().dtype(torch::kFloat32).device(torch::kCUDA);
+    auto o_part = torch::empty({p.batch, p.q_head, p.num_splits, p.head_dim}, fopt);
+    auto ml_part = torch::empty({p.batch, p.q_head, p.num_splits, 2}, fopt);
+    p.o_part = o_part.data_ptr<float>();
+    p.ml_part = ml_part.data_ptr<float>();
+
+    size_t smem = PDC_CHUNK * p.head_dim * sizeof(bf16);
+    paged_attn_decode_split_kv_kernel<<<
+        dim3(p.batch * p.kv_head, 1, p.num_splits),
+        dim3(32, group_size),
+        smem>>>(p);
+    paged_attn_decode_combine_kernel<<<p.batch * p.q_head, p.head_dim>>>(p);
+}
+
+#ifndef ASTRAI_NO_MMA
+template <int HEAD_DIM, int BC>
+static void launch_paged_mma_decode(PagedAttentionParams<bf16>& p) {
+    int tiles_total = (p.kv_len + BC - 1) / BC;
+    p.num_splits = paged_decode_num_splits(p.batch * p.kv_head, tiles_total);
+
+    auto fopt = torch::TensorOptions().dtype(torch::kFloat32).device(torch::kCUDA);
+    auto o_part = torch::empty({p.batch, p.q_head, p.num_splits, p.head_dim}, fopt);
+    auto ml_part = torch::empty({p.batch, p.q_head, p.num_splits, 2}, fopt);
+    p.o_part = o_part.data_ptr<float>();
+    p.ml_part = ml_part.data_ptr<float>();
+
+    paged_attn_decode_split_kv_mma_kernel<HEAD_DIM, BC>
+        <<<dim3(p.kv_head, p.batch, p.num_splits), 32>>>(p);
+    paged_attn_decode_combine_kernel<<<p.batch * p.q_head, p.head_dim>>>(p);
+}
+#endif
+
+template <int HEAD_DIM>
+static void dispatch_paged_decode(PagedAttentionParams<bf16>& p) {
+#ifndef ASTRAI_NO_MMA
+    int G = p.q_head / p.kv_head;
+    if (!p.use_mask && G >= 1 && G <= 16 && p.page_size >= 32) {
+        launch_paged_mma_decode<HEAD_DIM, 32>(p);
+        return;
+    }
+#endif
+    launch_paged_scalar_decode(p);
+}
+
+torch::Tensor attn_paged_decode(
+    torch::Tensor q,
+    torch::Tensor page_table,
+    torch::Tensor k_cache,
+    torch::Tensor v_cache,
+    int64_t page_size,
+    int64_t kv_len,
+    c10::optional<torch::Tensor> mask,
+    bool is_causal = false,
+    int64_t causal_offset = 0,
+    c10::optional<double> scale = c10::nullopt
+) {
+    const at::cuda::OptionalCUDAGuard device_guard(device_of(q));
+
+    int batch = q.size(0);
+    int q_head = q.size(1);
+    int head_dim = q.size(3);
+    int kv_head = k_cache.size(2);
+    int max_pages = page_table.size(1);
+
+    TORCH_CHECK(q.is_cuda() && page_table.is_cuda() && k_cache.is_cuda() && v_cache.is_cuda());
+    TORCH_CHECK(q.dtype() == torch::kBFloat16, "q must be bf16");
+    TORCH_CHECK(k_cache.dtype() == torch::kBFloat16, "k_cache must be bf16");
+    TORCH_CHECK(v_cache.dtype() == torch::kBFloat16, "v_cache must be bf16");
+    TORCH_CHECK(page_table.dtype() == torch::kLong, "page_table must be int64");
+    TORCH_CHECK(q.size(2) == 1, "Q seq_len must be 1 (decode)");
+    TORCH_CHECK(head_dim % 32 == 0, "head_dim must be multiple of 32");
+    TORCH_CHECK(k_cache.size(1) == page_size,
+                "k_cache dim 1 must equal page_size, got ",
+                k_cache.size(1), " vs ", page_size);
+    TORCH_CHECK(k_cache.size(0) >= 0, "k_cache must have at least 0 pages");
+
+    float scale_val = scale.has_value()
+        ? static_cast<float>(scale.value())
+        : 1.0f / std::sqrt(static_cast<float>(head_dim));
+
+    auto O = torch::empty_like(q);
+
+    PagedAttentionParams<bf16, float> p;
+    p.batch = batch;
+    p.q_head = q_head;
+    p.kv_head = kv_head;
+    p.q_len = 1;
+    p.kv_len = static_cast<int>(kv_len);
+    p.head_dim = head_dim;
+    p.use_mask = (mask.has_value() && mask.value().defined()) ? 1 : 0;
+    p.is_causal = is_causal ? 1 : 0;
+    p.causal_offset = static_cast<int>(causal_offset);
+    p.num_splits = 1;
+    p.scale = scale_val;
+    p.page_size = static_cast<int>(page_size);
+    p.max_pages = max_pages;
+    p.page_table = page_table.data_ptr<int64_t>();
+    p.k_cache = reinterpret_cast<const bf16*>(k_cache.data_ptr());
+    p.v_cache = reinterpret_cast<const bf16*>(v_cache.data_ptr());
+    p.q = reinterpret_cast<const bf16*>(q.data_ptr());
+    p.mask = p.use_mask ? mask.value().data_ptr<bool>() : nullptr;
+    p.o = reinterpret_cast<bf16*>(O.data_ptr());
+    p.o_part = nullptr;
+    p.ml_part = nullptr;
+
+    switch (p.head_dim) {
+        case 32:  dispatch_paged_decode<32>(p);  break;
+        case 64:  dispatch_paged_decode<64>(p);  break;
+        case 128: dispatch_paged_decode<128>(p); break;
+        case 256: dispatch_paged_decode<256>(p); break;
+        default:
+            TORCH_CHECK(false, "paged_decode: unsupported head_dim ", p.head_dim,
+                        " (supported: 32, 64, 128, 256)");
+    }
+
+    return O;
+}
+
+PYBIND11_MODULE(TORCH_EXTENSION_NAME, m) {
+    m.def("attn_paged_decode", &attn_paged_decode,
+        py::arg("q"),
+        py::arg("page_table"),
+        py::arg("k_cache"),
+        py::arg("v_cache"),
+        py::arg("page_size"),
+        py::arg("kv_len"),
+        py::arg("mask") = py::none(),
+        py::arg("is_causal") = false,
+        py::arg("causal_offset") = 0,
+        py::arg("scale") = py::none(),
+        "Paged GQA decode — split-KV with direct page-table access.");
+}

csrc/kernels/attn_paged_decode_split_kv.cuh

@@ -0,0 +1,140 @@
+#pragma once
+#include <cuda_bf16.h>
+#include <float.h>
+#include "attn_common.h"
+
+using bf16 = __nv_bfloat16;
+constexpr int PDC_CHUNK = 64;
+
+__device__ inline float paged_warp_reduce_sum(float val) {
+    for (int offset = 16; offset > 0; offset >>= 1)
+        val += __shfl_xor_sync(0xFFFFFFFF, val, offset);
+    return val;
+}
+
+// Split-KV scalar decode: one warp per query head, grid.z partitions KV.
+__global__ void paged_attn_decode_split_kv_kernel(PagedAttentionParams<bf16> p) {
+    int batch = blockIdx.x / p.kv_head;
+    int kv_head = blockIdx.x % p.kv_head;
+    int split = blockIdx.z;
+    int group_size = blockDim.y;
+    int q_head = kv_head * group_size + threadIdx.y;
+    int lane = threadIdx.x;
+    int hd_per_thread = p.head_dim / 32;
+
+    float q_reg[8];
+    int q_off = ((batch * p.q_head + q_head) * 1) * p.head_dim + lane * hd_per_thread;
+    #pragma unroll
+    for (int i = 0; i < hd_per_thread; i++)
+        q_reg[i] = __bfloat162float(p.q[q_off + i]);
+
+    float m = -FLT_MAX, d = 0.0f, acc_reg[8] = {0.0f};
+
+    extern __shared__ __align__(16) bf16 k_smem[];
+
+    int chunks_total = (p.kv_len + PDC_CHUNK - 1) / PDC_CHUNK;
+    int chunks_per_split = (chunks_total + p.num_splits - 1) / p.num_splits;
+    int ch_begin = split * chunks_per_split;
+    int ch_end = min(chunks_total, ch_begin + chunks_per_split);
+
+    const int mask_base = batch * p.kv_len;
+
+    for (int ci = ch_begin; ci < ch_end; ci++) {
+        int chunk_start = ci * PDC_CHUNK;
+        int this_chunk = min(PDC_CHUNK, p.kv_len - chunk_start);
+
+        int total = this_chunk * p.head_dim;
+        for (int i = threadIdx.y * 32 + lane; i < total; i += blockDim.x * blockDim.y) {
+            int s = i / p.head_dim;
+            int d_dim = i % p.head_dim;
+            int pos = chunk_start + s;
+            int logical_page = pos / p.page_size;
+            int page_offset = pos % p.page_size;
+            int phys_page = p.page_table[batch * p.max_pages + logical_page];
+            if (phys_page >= 0) {
+                int64_t off = (int64_t)phys_page * p.page_size * p.kv_head * p.head_dim
+                            + (int64_t)page_offset * p.kv_head * p.head_dim
+                            + (int64_t)kv_head * p.head_dim
+                            + d_dim;
+                k_smem[i] = p.k_cache[off];
+            } else {
+                k_smem[i] = __float2bfloat16(0.0f);
+            }
+        }
+        __syncthreads();
+
+        for (int s = 0; s < this_chunk; s++) {
+            float partial = 0.0f;
+            #pragma unroll
+            for (int i = 0; i < hd_per_thread; i++)
+                partial += q_reg[i] * __bfloat162float(k_smem[s * p.head_dim + lane * hd_per_thread + i]);
+            partial = paged_warp_reduce_sum(partial) * p.scale;
+
+            if (p.use_mask && p.mask && !p.mask[mask_base + chunk_start + s])
+                partial = -FLT_MAX;
+            if (p.is_causal && (chunk_start + s) > p.causal_offset)
+                partial = -FLT_MAX;
+
+            float new_m = fmaxf(m, partial);
+            float alpha = expf(m - new_m);
+            float beta  = expf(partial - new_m);
+            d = d * alpha + beta;
+
+            int pos = chunk_start + s;
+            int logical_page = pos / p.page_size;
+            int page_offset = pos % p.page_size;
+            int phys_page = p.page_table[batch * p.max_pages + logical_page];
+            if (phys_page >= 0) {
+                int64_t v_base = (int64_t)phys_page * p.page_size * p.kv_head * p.head_dim
+                               + (int64_t)page_offset * p.kv_head * p.head_dim
+                               + (int64_t)kv_head * p.head_dim;
+                #pragma unroll
+                for (int i = 0; i < hd_per_thread; i++)
+                    acc_reg[i] = acc_reg[i] * alpha + __bfloat162float(p.v_cache[v_base + lane * hd_per_thread + i]) * beta;
+            } else {
+                #pragma unroll
+                for (int i = 0; i < hd_per_thread; i++)
+                    acc_reg[i] = acc_reg[i] * alpha + 0.0f * beta;
+            }
+            m = new_m;
+        }
+        __syncthreads();
+    }
+
+    size_t bh = (size_t)batch * p.q_head + q_head;
+    size_t slot = bh * p.num_splits + split;
+    int d0 = lane * hd_per_thread;
+    #pragma unroll
+    for (int i = 0; i < hd_per_thread; i++)
+        p.o_part[slot * p.head_dim + (d0 + i)] = acc_reg[i];
+    if (lane == 0) {
+        p.ml_part[slot * 2] = m;
+        p.ml_part[slot * 2 + 1] = d;
+    }
+}
+
+__global__ void paged_attn_decode_combine_kernel(PagedAttentionParams<bf16> p) {
+    int bh = blockIdx.x;
+    int d = threadIdx.x;
+    if (d >= p.head_dim) return;
+
+    size_t split_base = (size_t)bh * p.num_splits;
+    const float* mlp = p.ml_part + split_base * 2;
+    const float* op = p.o_part + split_base * p.head_dim;
+
+    float m = -FLT_MAX, l = 0.0f, acc = 0.0f;
+    for (int s = 0; s < p.num_splits; s++) {
+        float mi = mlp[s * 2];
+        if (mi <= -FLT_MAX) continue;
+        float li = mlp[s * 2 + 1];
+        float nm = fmaxf(m, mi);
+        float corr = __expf(m - nm);
+        float e = __expf(mi - nm);
+        acc = acc * corr + op[s * p.head_dim + d] * e;
+        l = l * corr + li * e;
+        m = nm;
+    }
+
+    float inv = (l > 1e-20f) ? (1.0f / l) : 0.0f;
+    p.o[(size_t)bh * p.head_dim + d] = __float2bfloat16(acc * inv);
+}

csrc/kernels/attn_paged_decode_split_kv_mma.cuh

@@ -0,0 +1,174 @@
+#pragma once
+#include <cfloat>
+#include <cuda_bf16.h>
+#include "attn_common.h"
+#include "attn_mma_utils.cuh"
+
+using bf16 = __nv_bfloat16;
+
+// Paged split-KV tensor-core decode via GQA head-packing.
+// Identical algorithm to attn_decode_split_kv_mma_kernel but reads K/V
+// directly from the page pool through a page table, eliminating the gather
+// copy.  Each tile (BC=32) fits within a single page (page_size >= 32), so
+// the page-table lookup happens once per tile for cp.async.
+template <int HEAD_DIM, int BC>
+__global__ void paged_attn_decode_split_kv_mma_kernel(PagedAttentionParams<bf16> p) {
+    constexpr int BR = 16;
+    constexpr int KD = HEAD_DIM / 16;
+    constexpr int NC8 = BC / 8;
+    constexpr int KT2 = BC / 16;
+    constexpr int DN8 = HEAD_DIM / 8;
+    constexpr int LD = HEAD_DIM;
+    constexpr int SWIZ_MASK = (HEAD_DIM >= 64) ? 7 : (HEAD_DIM / 8 - 1);
+
+    const int lane = threadIdx.x;
+    const int gid = lane >> 2;
+    const int tid4 = lane & 3;
+
+    const int kv_head_idx = blockIdx.x;
+    const int batch = blockIdx.y;
+    const int split = blockIdx.z;
+    const int G = p.q_head / p.kv_head;
+    const int q_head0 = kv_head_idx * G;
+
+    __shared__ __align__(16) bf16 sK[BC * HEAD_DIM];
+    __shared__ __align__(16) bf16 sV[BC * HEAD_DIM];
+    __shared__ __align__(16) bf16 sQ[BR * HEAD_DIM];
+
+    // ---- load Q into registers via ldmatrix ----
+    for (int i = lane; i < BR * HEAD_DIM; i += 32) {
+        int r = i / HEAD_DIM, d = i % HEAD_DIM;
+        bf16 val = __float2bfloat16(0.0f);
+        if (r < G) {
+            int qh = q_head0 + r;
+            val = p.q[(batch * p.q_head + qh) * HEAD_DIM + d];
+        }
+        sQ[r * LD + swiz_col(d, r, SWIZ_MASK)] = val;
+    }
+    __syncwarp();
+
+    unsigned Qa[KD][4];
+    int qrow_l = (lane & 7) + (lane & 8);
+    int qcol_l = (lane & 16) ? 8 : 0;
+    #pragma unroll
+    for (int kt = 0; kt < KD; kt++)
+        ldmatrix_x4(Qa[kt], &sQ[qrow_l * LD + swiz_col(kt * 16 + qcol_l, qrow_l, SWIZ_MASK)]);
+
+    float Oacc[DN8][4];
+    #pragma unroll
+    for (int j = 0; j < DN8; j++)
+        Oacc[j][0] = Oacc[j][1] = Oacc[j][2] = Oacc[j][3] = 0.0f;
+    float m0 = -FLT_MAX, m1 = -FLT_MAX, l0 = 0.0f, l1 = 0.0f;
+
+    const int mask_base = batch * p.kv_len;
+    const int tiles_total = (p.kv_len + BC - 1) / BC;
+    const int tiles_per_split = (tiles_total + p.num_splits - 1) / p.num_splits;
+    const int ti_begin = split * tiles_per_split;
+    const int ti_end = min(tiles_total, ti_begin + tiles_per_split);
+    const int has_mask = p.use_mask && p.mask;
+
+    // Paged strides (constant for the block)
+    const int64_t page_stride = (int64_t)p.page_size * p.kv_head * HEAD_DIM;
+    const int64_t pos_stride  = (int64_t)p.kv_head * HEAD_DIM;
+    const int64_t head_off    = (int64_t)kv_head_idx * HEAD_DIM;
+
+    for (int ti = ti_begin; ti < ti_end; ti++) {
+        int kv0 = ti * BC;
+
+        // phys_page is constant for the whole tile (BC <= page_size).
+        int logical_page = kv0 / p.page_size;
+        int phys_page = p.page_table[batch * p.max_pages + logical_page];
+        bool page_valid = (phys_page >= 0);
+
+        bool full_tile = page_valid && (kv0 + BC <= p.kv_len);
+        if (full_tile) {
+            constexpr int VEC = 8;
+            int total = BC * HEAD_DIM;
+            #pragma unroll
+            for (int i = lane * VEC; i < total; i += 32 * VEC) {
+                int r = i / HEAD_DIM, d = i % HEAD_DIM;
+                int kc = kv0 + r;
+                int page_off = kc % p.page_size;
+                int64_t gmem_base = (int64_t)phys_page * page_stride
+                                  + (int64_t)page_off * pos_stride
+                                  + head_off;
+                cp_async_16(&sK[r * LD + swiz_col(d, r, SWIZ_MASK)],
+                            &p.k_cache[gmem_base + d]);
+                cp_async_16(&sV[r * LD + swiz_col(d, r, SWIZ_MASK)],
+                            &p.v_cache[gmem_base + d]);
+            }
+            cp_async_commit();
+            cp_async_wait_all();
+        } else {
+            for (int i = lane; i < BC * HEAD_DIM; i += 32) {
+                int r = i / HEAD_DIM, d = i % HEAD_DIM;
+                int kc = kv0 + r;
+                bf16 z = __float2bfloat16(0.0f);
+                if (kc < p.kv_len && page_valid) {
+                    int page_off = kc % p.page_size;
+                    int64_t gmem_base = (int64_t)phys_page * page_stride
+                                      + (int64_t)page_off * pos_stride
+                                      + head_off;
+                    sK[r * LD + swiz_col(d, r, SWIZ_MASK)] = p.k_cache[gmem_base + d];
+                    sV[r * LD + swiz_col(d, r, SWIZ_MASK)] = p.v_cache[gmem_base + d];
+                } else {
+                    sK[r * LD + swiz_col(d, r, SWIZ_MASK)] = z;
+                    sV[r * LD + swiz_col(d, r, SWIZ_MASK)] = z;
+                }
+            }
+        }
+        __syncwarp();
+
+        float Sacc[NC8][4];
+        mma_compute_scores<KD, NC8>(Qa, sK, LD, SWIZ_MASK, lane, Sacc);
+
+        #pragma unroll
+        for (int n8 = 0; n8 < NC8; n8++)
+            Sacc[n8][0] *= p.scale, Sacc[n8][1] *= p.scale,
+            Sacc[n8][2] *= p.scale, Sacc[n8][3] *= p.scale;
+
+        int maxc = p.is_causal ? min(p.kv_len, p.causal_offset + 1) : p.kv_len;
+        mma_softmax_tile<NC8, DN8>(kv0, maxc, maxc,
+                                    mask_base, p.mask, has_mask,
+                                    Sacc, Oacc, m0, m1, l0, l1, lane);
+
+        mma_pv_accumulate<DN8, KT2>(Sacc, sV, LD, SWIZ_MASK, lane, Oacc);
+        __syncwarp();
+    }
+
+    // ---- write UN-normalised partials for this split ----
+    auto split_slot = [&](int h) -> size_t {
+        size_t bh = (size_t)batch * p.q_head + h;
+        return bh * p.num_splits + split;
+    };
+    #pragma unroll
+    for (int dn8 = 0; dn8 < DN8; dn8++) {
+        int d = dn8 * 8 + 2 * tid4;
+        int r0 = gid, r1 = gid + 8;
+        if (r0 < G) {
+            int h = q_head0 + r0;
+            float* op = p.o_part + split_slot(h) * HEAD_DIM;
+            op[d] = Oacc[dn8][0];
+            op[d + 1] = Oacc[dn8][1];
+        }
+        if (r1 < G) {
+            int h = q_head0 + r1;
+            float* op = p.o_part + split_slot(h) * HEAD_DIM;
+            op[d] = Oacc[dn8][2];
+            op[d + 1] = Oacc[dn8][3];
+        }
+    }
+    if (tid4 == 0) {
+        int r0 = gid, r1 = gid + 8;
+        if (r0 < G) {
+            int h = q_head0 + r0;
+            float* mp = p.ml_part + split_slot(h) * 2;
+            mp[0] = m0; mp[1] = l0;
+        }
+        if (r1 < G) {
+            int h = q_head0 + r1;
+            float* mp = p.ml_part + split_slot(h) * 2;
+            mp[0] = m1; mp[1] = l1;
+        }
+    }
+}

csrc/kernels/attn_prefill.cu

@@ -0,0 +1,81 @@
+#include "attn_prefill_split_q.cuh"
+#include "attn_entry_utils.cuh"
+
+#ifndef ASTRAI_NO_MMA
+#include "attn_prefill_split_q_mma.cuh"
+#endif
+
+template <int HEAD_DIM>
+static void dispatch_prefill(AttentionParams<bf16>& p) {
+#ifndef ASTRAI_NO_MMA
+    constexpr int WARPS = 4, BR = 16;
+    // KV tile: bigger tiles amortize the per-tile cp.async wait + barrier +
+    // loop overhead over more tensor-core work (this kernel is latency-bound,
+    // not compute/bandwidth-bound), so BC=32 wins ~6-8% over BC=16 for
+    // D<=128. D=256 stays at 16: BC=32 double-buffered would need 64KB smem,
+    // over the 48KB static cap. Both keep 3 blocks/SM (2 for D=256).
+    constexpr int BC = (HEAD_DIM <= 128) ? 32 : 16;
+    // Register-hint MIN_BLOCKS tuned per HEAD_DIM's (BC=32) smem+register
+    // footprint: the largest blocks/SM that avoids register spills.
+    constexpr int MIN_BLOCKS = (HEAD_DIM <= 32) ? 6 : (HEAD_DIM <= 64) ? 4
+                             : (HEAD_DIM <= 128) ? 3 : 2;
+    dim3 grid((p.q_len + BR * WARPS - 1) / (BR * WARPS), p.q_head, p.batch);
+    dim3 block(WARPS * 32, 1, 1);
+    // Static shared memory — no dynamic smem or cudaFuncSetAttribute needed.
+    // sK[BC*LD] + sV[BC*LD] + sQ[BR*LD], all sized by template params.
+    attn_prefill_split_q_mma_kernel<HEAD_DIM, WARPS, BC, MIN_BLOCKS><<<grid, block>>>(p);
+#else
+    constexpr int G = 8, ROWS = 32, P_BC = 32;
+    dim3 grid((p.q_len + ROWS - 1) / ROWS, p.q_head, p.batch);
+    dim3 block(G, ROWS, 1);
+    attn_prefill_split_q_kernel_t<HEAD_DIM, G, ROWS, P_BC><<<grid, block>>>(p);
+#endif
+}
+
+torch::Tensor attn_prefill(
+    torch::Tensor q,
+    torch::Tensor k,
+    torch::Tensor v,
+    c10::optional<torch::Tensor> mask,
+    bool is_causal = false,
+    int64_t causal_offset = 0,
+    c10::optional<double> scale = c10::nullopt
+) {
+    AttentionParams<bf16> p;
+    attn_pack_params(q, k, v, mask, is_causal, causal_offset, scale, p);
+    TORCH_CHECK(p.head_dim % 16 == 0, "head_dim must be multiple of 16");
+
+    auto O = torch::empty_like(q);
+    p.o = (bf16*)O.data_ptr();
+
+    switch (p.head_dim) {
+        case 32:
+            dispatch_prefill<32>(p);
+            break;
+        case 64:
+            dispatch_prefill<64>(p);
+            break;
+        case 128:
+            dispatch_prefill<128>(p);
+            break;
+        case 256:
+            dispatch_prefill<256>(p);
+            break;
+        default:
+            TORCH_CHECK(false, "prefill: unsupported head_dim ", p.head_dim,
+                        " (supported: 32,64,128,256)");
+    }
+    return O;
+}
+
+PYBIND11_MODULE(TORCH_EXTENSION_NAME, m) {
+    m.def("attn_prefill", &attn_prefill,
+        py::arg("q"),
+        py::arg("k"),
+        py::arg("v"),
+        py::arg("mask") = py::none(),
+        py::arg("is_causal") = false,
+        py::arg("causal_offset") = 0,
+        py::arg("scale") = py::none(),
+        "GQA prefill (tensor-core mma on sm_80+, scalar fallback)");
+}

csrc/kernels/attn_prefill_split_q.cuh

@@ -0,0 +1,140 @@
+#pragma once
+#include <cfloat>
+#include <cuda_bf16.h>
+#include "attn_common.h"
+
+using bf16 = __nv_bfloat16;
+
+// v9: group-split register blocking. G threads cooperate on one query row,
+// each owning HEAD_DIM/G dims of qreg[]/acc[]. Small per-thread footprint keeps
+// occupancy high; the S dot product is reduced across the G-lane group with a
+// short shuffle chain (log2(G) shuffles) instead of a full 32-lane warp reduce.
+// Online (per-kv) softmax — cheap because acc[] is only HEAD_DIM/G long.
+// Templated on <HEAD_DIM, G, ROWS, P_BC>. Block = (G, ROWS). G power-of-two,
+// G*ROWS a multiple of 32 with groups warp-aligned.
+
+template <int G>
+__device__ __forceinline__ float group_reduce_sum(float v, unsigned mask) {
+#pragma unroll
+    for (int o = G / 2; o > 0; o >>= 1)
+        v += __shfl_xor_sync(mask, v, o);
+    return v;
+}
+
+// load 8 contiguous bf16 from (16-byte aligned) smem as one float4, unpack to
+// 8 floats — cuts shared-load instructions 8x vs scalar bf16 loads.
+__device__ __forceinline__ void ld8(const bf16* p, float* o) {
+    float4 raw = *reinterpret_cast<const float4*>(p);
+    const __nv_bfloat162* h = reinterpret_cast<const __nv_bfloat162*>(&raw);
+#pragma unroll
+    for (int j = 0; j < 4; j++) {
+        float2 f = __bfloat1622float2(h[j]);
+        o[2 * j] = f.x;
+        o[2 * j + 1] = f.y;
+    }
+}
+
+template <int HEAD_DIM, int G, int ROWS, int P_BC>
+__global__ void attn_prefill_split_q_kernel_t(AttentionParams<bf16> p) {
+    constexpr int DPT = HEAD_DIM / G;
+
+    int q_tile = blockIdx.x;
+    int q_head = blockIdx.y;
+    int batch  = blockIdx.z;
+    int gpos   = threadIdx.x;  // 0..G-1  (which d-chunk)
+    int row    = threadIdx.y;  // 0..ROWS-1
+    int q_row  = q_tile * ROWS + row;
+
+    int kv_head = q_head / (p.q_head / p.kv_head);
+
+    __shared__ __align__(16) bf16 sK[P_BC * HEAD_DIM];
+    __shared__ __align__(16) bf16 sV[P_BC * HEAD_DIM];
+
+    float qreg[DPT];
+    if (q_row < p.q_len) {
+        int q_off = ((batch * p.q_head + q_head) * p.q_len + q_row) * HEAD_DIM + gpos * DPT;
+#pragma unroll
+        for (int i = 0; i < DPT; i++)
+            qreg[i] = __bfloat162float(p.q[q_off + i]) * p.scale;
+    }
+
+    float m = -FLT_MAX, l = 0.0f;
+    float acc[DPT];
+#pragma unroll
+    for (int i = 0; i < DPT; i++)
+        acc[i] = 0.0f;
+
+    int kv_base = ((batch * p.kv_head + kv_head) * p.kv_len) * HEAD_DIM;
+    int tiles   = (p.kv_len + P_BC - 1) / P_BC;
+    int tt      = G * ROWS;
+    int lid     = row * G + gpos;
+
+    // per-group shuffle mask: only the G lanes of this row's group participate,
+    // so causal masking (differing loop bounds across rows in a warp) is safe.
+    int lane_in_warp = lid & 31;
+    unsigned gmask = (G == 32) ? 0xFFFFFFFFu
+                               : (((1u << G) - 1u) << (lane_in_warp & ~(G - 1)));
+
+    for (int ti = 0; ti < tiles; ti++) {
+        int kv0  = ti * P_BC;
+        int tlen = min(P_BC, p.kv_len - kv0);
+
+        for (int i = lid; i < tlen * HEAD_DIM; i += tt) {
+            int gidx = kv_base + (kv0 + i / HEAD_DIM) * HEAD_DIM + (i % HEAD_DIM);
+            sK[i] = p.k[gidx];
+            sV[i] = p.v[gidx];
+        }
+        __syncthreads();
+
+        int lim = tlen;
+        if (p.is_causal && q_row < p.q_len) {
+            int ep = q_row + p.causal_offset + 1;
+            if (kv0 >= ep)
+                lim = 0;
+            else if (kv0 + tlen > ep)
+                lim = ep - kv0;
+        }
+
+        for (int s = 0; s < lim; s++) {
+            const bf16* kr = sK + s * HEAD_DIM + gpos * DPT;
+            float part = 0.0f;
+#pragma unroll
+            for (int i = 0; i < DPT; i += 8) {
+                float k8[8];
+                ld8(kr + i, k8);
+#pragma unroll
+                for (int j = 0; j < 8; j++)
+                    part = fmaf(qreg[i + j], k8[j], part);
+            }
+            float dot = group_reduce_sum<G>(part, gmask);
+
+            if (p.use_mask && p.mask && !p.mask[batch * p.kv_len + kv0 + s])
+                dot = -FLT_MAX;
+
+            float nm = fmaxf(m, dot);
+            float al = __expf(m - nm);
+            float be = __expf(dot - nm);
+            l = l * al + be;
+
+            const bf16* vr = sV + s * HEAD_DIM + gpos * DPT;
+#pragma unroll
+            for (int i = 0; i < DPT; i += 8) {
+                float v8[8];
+                ld8(vr + i, v8);
+#pragma unroll
+                for (int j = 0; j < 8; j++)
+                    acc[i + j] = fmaf(v8[j], be, acc[i + j] * al);
+            }
+            m = nm;
+        }
+        __syncthreads();
+    }
+
+    if (q_row < p.q_len) {
+        int o_off = ((batch * p.q_head + q_head) * p.q_len + q_row) * HEAD_DIM + gpos * DPT;
+        float rl = (l > 1e-10f) ? (1.0f / l) : 0.0f;
+#pragma unroll
+        for (int i = 0; i < DPT; i++)
+            p.o[o_off + i] = __float2bfloat16(acc[i] * rl);
+    }
+}

csrc/kernels/attn_prefill_split_q_mma.cuh

@@ -0,0 +1,209 @@
+#pragma once
+#include <cfloat>
+#include <cuda_bf16.h>
+#include "attn_common.h"
+#include "attn_mma_utils.cuh"
+
+using bf16 = __nv_bfloat16;
+
+// Tensor-core prefill flash attention (raw mma.sync PTX).
+// One warp owns BR=16 query rows. S = Q@K^T and O = P@V run on bf16 tensor
+// cores via mma.sync.m16n8k16 (f32 accumulate). Q fragments are loaded once
+// straight from global into the mma A-operand layout (no smem staging) and
+// kept resident in registers across the tile loop. S, O, and the online-softmax
+// stats (m, l) also live in registers.
+// Shared memory is statically sized via template parameters — no dynamic
+// allocation. The mma fragment layout is used directly: the S accumulator
+// (f32) maps element-for-element onto the P matrix_a (bf16) operand, so
+// softmax needs no shuffle repack; row reductions fold across the 4-lane
+// thread group. Templated on <HEAD_DIM, WARPS, BC, MIN_BLOCKS> with BC a
+// multiple of 16.
+//
+// Occupancy: __launch_bounds__ forces the compiler to fit MIN_BLOCKS blocks/SM,
+// spilling to local memory as needed. MIN_BLOCKS is tuned per HEAD_DIM to the
+// double-buffered smem footprint (2*BC*LD for each of K/V).
+//
+// Software pipeline: K/V are double-buffered and loaded via cp.async one tile
+// ahead, so the next tile streams from global memory while the current tile's
+// tensor-core math runs — hiding load latency (long_scoreboard). A single
+// __syncthreads per tile both publishes the freshly loaded tile cross-warp and
+// (because it runs before the next prefetch) guards the buffer being refilled,
+// so no second barrier is needed. Predicated cp.async (cp_async_16_pred)
+// zero-fills rows past kv_len, unifying full and partial tiles on one path.
+// BC=32 (D<=128) amortizes the per-tile wait+barrier+loop overhead over more
+// tensor-core work — this kernel is latency-bound (low occupancy from high
+// register pressure), so fewer, larger tiles beat many tiny ones.
+//
+// Optimizations: load Q fragments directly from global in mma A-operand layout
+// (no sQ staging, no prologue barriers); pre-scale Q by attention scale during Q load; packed bf16x2 output stores;
+// causal tile skipping (block-level prefetch bound + warp-level compute skip);
+// XOR swizzle (swiz_col) → eliminates ldmatrix bank conflicts without LD
+// padding (LD=HEAD_DIM).
+
+template <int HEAD_DIM, int WARPS, int BC, int MIN_BLOCKS>
+__global__ __launch_bounds__(WARPS * 32, MIN_BLOCKS)
+void attn_prefill_split_q_mma_kernel(AttentionParams<bf16> p) {
+    constexpr int BR = 16;
+    constexpr int KD = HEAD_DIM / 16;  // Q/K k-tiles
+    constexpr int NC8 = BC / 8;        // S n-tiles (N=8 each)
+    constexpr int KT2 = BC / 16;       // P k-tiles (K=16 each)
+    constexpr int DN8 = HEAD_DIM / 8;  // O n-tiles (N=8 each)
+    constexpr int LD = HEAD_DIM;   // XOR swizzle (swiz_col) handles bank conflicts
+    constexpr int SWIZ_MASK = (HEAD_DIM >= 64) ? 7 : (HEAD_DIM / 8 - 1);  // chunk bits, stay within LD
+
+    const int warp = threadIdx.x / 32;
+    const int lane = threadIdx.x % 32;
+    const int gid = lane >> 2;   // 0..7  → rows gid, gid+8
+    const int tid4 = lane & 3;   // 0..3
+    const int nthreads = WARPS * 32;
+
+    const int q_head = blockIdx.y;
+    const int batch = blockIdx.z;
+    const int kv_head = q_head / (p.q_head / p.kv_head);
+    const int qrow0 = (blockIdx.x * WARPS + warp) * BR;
+
+    // Static shared memory — sized by template parameters at compile time.
+    // K/V are double-buffered (STAGES=2): the next tile's cp.async load runs
+    // while the current tile's tensor-core math executes, hiding global-load
+    // latency (FA2-style software pipeline). No dynamic smem / carveout opt-in.
+    constexpr int STAGES = 2;
+    __shared__ __align__(16) bf16 sK[STAGES * BC * LD];
+    __shared__ __align__(16) bf16 sV[STAGES * BC * LD];
+
+    // Load the Q fragments straight from global into the mma A-operand layout
+    // (m16n8k16, row-major): no sQ staging area and no serialized per-warp
+    // prologue barriers. Each lane reads exactly the 8 Q elements ldmatrix
+    // would have produced, pre-scaled by the attention scale. Kept resident in
+    // registers across the tile loop.
+    //   frag[0]/[2]: row = qrow0 + gid ;  frag[1]/[3]: row = qrow0 + gid + 8
+    //   frag[0]/[1]: cols kt*16 + tid4*2 + {0,1} ; frag[2]/[3]: + 8
+    const int q_base = ((batch * p.q_head + q_head) * p.q_len) * HEAD_DIM;
+    const int qra = qrow0 + gid;
+    const int qrb = qrow0 + gid + 8;
+    const bool va = qra < p.q_len, vb = qrb < p.q_len;
+    unsigned Qa[KD][4];
+#pragma unroll
+    for (int kt = 0; kt < KD; kt++) {
+        int c = kt * 16 + tid4 * 2;
+        const unsigned* pau = reinterpret_cast<const unsigned*>(
+            &p.q[q_base + qra * HEAD_DIM + c]);
+        const unsigned* pbu = reinterpret_cast<const unsigned*>(
+            &p.q[q_base + qrb * HEAD_DIM + c]);
+        Qa[kt][0] = va ? pau[0] : 0u;
+        Qa[kt][1] = vb ? pbu[0] : 0u;
+        Qa[kt][2] = va ? pau[4] : 0u;
+        Qa[kt][3] = vb ? pbu[4] : 0u;
+    }
+
+    float Oacc[DN8][4];
+#pragma unroll
+    for (int j = 0; j < DN8; j++)
+        Oacc[j][0] = Oacc[j][1] = Oacc[j][2] = Oacc[j][3] = 0.0f;
+    float m0 = -FLT_MAX, m1 = -FLT_MAX, l0 = 0.0f, l1 = 0.0f;
+
+    const int kv_base = ((batch * p.kv_head + kv_head) * p.kv_len) * HEAD_DIM;
+    const int tiles = (p.kv_len + BC - 1) / BC;
+    const int qr0 = qrow0 + gid;      // row for c0/c1
+    const int qr1 = qrow0 + gid + 8;  // row for c2/c3
+
+    // Causal tile-skip bounds (no-op when is_causal == 0)
+    const int use_skip = p.is_causal;
+    const int max_kv = qrow0 + BR - 1 + p.causal_offset;
+    const int block_max_kv =
+        blockIdx.x * WARPS * BR + WARPS * BR - 1 + p.causal_offset;
+    const int has_mask = p.use_mask && p.mask;
+    const int mb = batch * p.kv_len;
+
+    // Last active tile: block-level causal bound (all warps in the block share
+    // the K/V load, so the prefetch range is the block max, not per-warp).
+    int t_end = tiles - 1;
+    if (use_skip) {
+        int bt = block_max_kv / BC;
+        if (bt < t_end) t_end = bt;
+    }
+
+    constexpr int VEC = 8;  // bf16 per cp.async unit (16 bytes)
+    constexpr int TOTAL = BC * HEAD_DIM;
+
+    // Issue cp.async loads for tile `ti` into shared buffer `buf`. Predicated
+    // loads zero-fill rows past kv_len, so partial tiles need no scalar path.
+    auto load_tile = [&](int ti, int buf) {
+        int kv0 = ti * BC;
+        bf16* dK = sK + buf * BC * LD;
+        bf16* dV = sV + buf * BC * LD;
+#pragma unroll
+        for (int i = threadIdx.x * VEC; i < TOTAL; i += nthreads * VEC) {
+            int r = i / HEAD_DIM, d = i % HEAD_DIM;
+            int kc = kv0 + r;
+            bool valid = kc < p.kv_len;
+            int off = r * LD + swiz_col(d, r, SWIZ_MASK);
+            cp_async_16_pred(&dK[off], &p.k[kv_base + kc * HEAD_DIM + d], valid);
+            cp_async_16_pred(&dV[off], &p.v[kv_base + kc * HEAD_DIM + d], valid);
+        }
+        cp_async_commit();
+    };
+
+    // Prologue: kick off the first tile's load.
+    load_tile(0, 0);
+
+    for (int ti = 0; ti <= t_end; ti++) {
+        int buf = ti & 1;
+
+        // Wait for the current tile's async copies, then a single barrier: it
+        // both publishes this tile's data cross-warp AND guarantees the prior
+        // compute on the buffer we are about to refill has finished. Issuing
+        // the next tile's load *after* this barrier lets one barrier cover both
+        // hazards (vs two), while the load still overlaps this tile's math.
+        cp_async_wait_group<0>();
+        __syncthreads();
+        if (ti < t_end) load_tile(ti + 1, (ti + 1) & 1);
+
+        const bf16* bK = sK + buf * BC * LD;
+        const bf16* bV = sV + buf * BC * LD;
+        int kv0 = ti * BC;
+
+        // Warp-level causal skip
+        if (!use_skip || kv0 <= max_kv) {
+
+        // S = Q @ K^T + scale + online softmax + O += P @ V
+        float Sacc[NC8][4];
+        mma_compute_scores<KD, NC8>(Qa, bK, LD, SWIZ_MASK, lane, Sacc);
+
+        // post-multiply scale in float (no bf16 precision loss from pre-scaling Q)
+        #pragma unroll
+        for (int n8 = 0; n8 < NC8; n8++)
+            Sacc[n8][0] *= p.scale, Sacc[n8][1] *= p.scale,
+            Sacc[n8][2] *= p.scale, Sacc[n8][3] *= p.scale;
+
+        int maxc0 = p.is_causal ? min(p.kv_len, qr0 + p.causal_offset + 1)
+                                : p.kv_len;
+        int maxc1 = p.is_causal ? min(p.kv_len, qr1 + p.causal_offset + 1)
+                                : p.kv_len;
+        mma_softmax_tile<NC8, DN8>(kv0, maxc0, maxc1,
+                                    mb, p.mask, has_mask,
+                                    Sacc, Oacc, m0, m1, l0, l1, lane);
+
+        mma_pv_accumulate<DN8, KT2>(Sacc, bV, LD, SWIZ_MASK, lane, Oacc);
+        }  // if active (warp-level causal skip)
+    }
+
+    // ---- write output ---- (packed bf16x2 stores: one 32-bit STG per pair,
+    // halves store count and removes the uncoalesced scalar-store penalty)
+    float rl0 = (l0 > 1e-20f) ? (1.0f / l0) : 0.0f;
+    float rl1 = (l1 > 1e-20f) ? (1.0f / l1) : 0.0f;
+    const int o_base = ((batch * p.q_head + q_head) * p.q_len) * HEAD_DIM;
+#pragma unroll
+    for (int dn8 = 0; dn8 < DN8; dn8++) {
+        int d = dn8 * 8 + 2 * tid4;
+        if (qr0 < p.q_len) {
+            __nv_bfloat162 v = __floats2bfloat162_rn(Oacc[dn8][0] * rl0,
+                                                     Oacc[dn8][1] * rl0);
+            *reinterpret_cast<__nv_bfloat162*>(&p.o[o_base + qr0 * HEAD_DIM + d]) = v;
+        }
+        if (qr1 < p.q_len) {
+            __nv_bfloat162 v = __floats2bfloat162_rn(Oacc[dn8][2] * rl1,
+                                                     Oacc[dn8][3] * rl1);
+            *reinterpret_cast<__nv_bfloat162*>(&p.o[o_base + qr1 * HEAD_DIM + d]) = v;
+        }
+    }
+}

csrc/tests/attn_decode_test.cu

@@ -0,0 +1,225 @@
+/*
+Pure-C test:
+nvcc -I csrc -arch=sm_89 -O3 \
+    --use_fast_math --ptxas-options=-O3 --extra-device-vectorization \
+    csrc/tests/attn_decode_test.cu -o test && ./test
+*/
+
+#include "test_utils.cuh"
+#include "../kernels/attn_decode_split_kv.cuh"
+#ifndef ASTRAI_NO_MMA
+#include "../kernels/attn_decode_split_kv_mma.cuh"
+#endif
+
+// Split-K scratch (torch-free): the production launcher allocates these from
+// torch; here we pass pre-allocated device buffers so the bench loop doesn't
+// pay a cudaMalloc per iteration. Size for the maximum split count (32).
+struct DecodeScratch {
+    float* o_part = nullptr;
+    float* ml_part = nullptr;
+};
+
+// Launch the production decode path (tensor-core head-packing MMA on sm_80+,
+// scalar fallback otherwise), mirroring dispatch_decode() in attn_decode.cu.
+#ifndef ASTRAI_NO_MMA
+static bool decode_use_mma(const AttentionParams<bf16>& p) {
+    int G = p.q_head / p.kv_head;
+    return !p.use_mask && G > 1 && G <= 16;
+}
+
+template <int HEAD_DIM, int BC>
+static void launch_mma_decode(AttentionParams<bf16>& p, DecodeScratch& sc) {
+    int tiles_total = (p.kv_len + BC - 1) / BC;
+    p.num_splits = compute_num_splits(p.batch * p.kv_head, tiles_total);
+    p.o_part = sc.o_part;
+    p.ml_part = sc.ml_part;
+
+    attn_decode_split_kv_mma_kernel<HEAD_DIM, BC>
+        <<<dim3(p.kv_head, p.batch, p.num_splits), 32>>>(p);
+    attn_decode_combine_kernel<<<p.batch * p.q_head, p.head_dim>>>(p);
+}
+#endif
+
+static void launch_scalar_decode(AttentionParams<bf16>& p, DecodeScratch& sc) {
+    int gs = p.q_head / p.kv_head;
+    int chunks_total = (p.kv_len + DC_CHUNK - 1) / DC_CHUNK;
+    p.num_splits = compute_num_splits(p.batch * p.kv_head, chunks_total);
+    p.o_part = sc.o_part;
+    p.ml_part = sc.ml_part;
+
+    size_t smem = DC_CHUNK * p.head_dim * sizeof(bf16);
+    attn_decode_split_kv_kernel<<<dim3(p.batch * p.kv_head, 1, p.num_splits), dim3(32, gs), smem>>>(p);
+    attn_decode_combine_kernel<<<p.batch * p.q_head, p.head_dim>>>(p);
+}
+
+template <int HEAD_DIM>
+static void dispatch_decode_t(AttentionParams<bf16>& p, DecodeScratch& sc) {
+#ifndef ASTRAI_NO_MMA
+    if (decode_use_mma(p)) { launch_mma_decode<HEAD_DIM, 32>(p, sc); return; }
+#endif
+    launch_scalar_decode(p, sc);
+}
+
+static void dispatch_decode(AttentionParams<bf16>& p, DecodeScratch& sc) {
+    switch (p.head_dim) {
+        case 32:  dispatch_decode_t<32>(p, sc);  break;
+        case 64:  dispatch_decode_t<64>(p, sc);  break;
+        case 128: dispatch_decode_t<128>(p, sc); break;
+        case 256: dispatch_decode_t<256>(p, sc); break;
+        default:  printf("bench: unsupported D=%d\n", p.head_dim);
+    }
+}
+
+// Warmed-up, CUDA-event timed sweep over the production decode MMA path.
+// Decode (q_len==1) is memory-bound: the two matmuls are GEMV-shaped, so we
+// report both effective K/V read bandwidth and the (small) attention FLOP/s.
+// FLOP/s = 2 matmuls (q@K^T, P@V), each 2*B*Hq*kv*D flops.
+// Bytes    = K + V read = 2 * B*Hk*kv*D * sizeof(bf16).
+static void bench() {
+    const int cfgs[][5] = {
+        {1, 32, 4, 512, 128},    // B,Hq,Hk,seq,D
+        {1, 32, 4, 1024, 128},
+        {1, 32, 4, 2048, 128},
+        {1, 32, 4, 4096, 128},
+        {16, 32, 4, 2048, 128},
+        {32, 32, 4, 1024, 128},
+    };
+    int n = sizeof(cfgs)/sizeof(cfgs[0]);
+    const int WARMUP = 10, ITERS = 100;
+    printf("\n===== DECODE BENCH (warmup=%d iters=%d) =====\n", WARMUP, ITERS);
+    printf("%-46s | %10s | %10s | %10s\n",
+           "config", "latency", "bandwidth", "throughput");
+    printf("---------------------------------------------------------------"
+           "----------------------------\n");
+
+    for (int ci = 0; ci < n; ci++) {
+        int B=cfgs[ci][0], Hq=cfgs[ci][1], Hk=cfgs[ci][2];
+        int sl=cfgs[ci][3], D=cfgs[ci][4];
+        size_t nQ=(size_t)B*Hq*D, nKV=(size_t)B*Hk*sl*D;
+
+        bf16 *dQ,*dK,*dV,*dO,*tmp;
+        cudaMalloc(&dQ,nQ*2); cudaMalloc(&dK,nKV*2);
+        cudaMalloc(&dV,nKV*2); cudaMalloc(&dO,nQ*2);
+        size_t big = nQ>nKV?nQ:nKV; tmp=new bf16[big];
+        for (size_t i=0;i<nQ;i++)  tmp[i]=f2bf(randf());
+        cudaMemcpy(dQ,tmp,nQ*2,cudaMemcpyHostToDevice);
+        for (size_t i=0;i<nKV;i++) tmp[i]=f2bf(randf());
+        cudaMemcpy(dK,tmp,nKV*2,cudaMemcpyHostToDevice);
+        for (size_t i=0;i<nKV;i++) tmp[i]=f2bf(randf());
+        cudaMemcpy(dV,tmp,nKV*2,cudaMemcpyHostToDevice);
+
+        AttentionParams<bf16> p;
+        p.batch=B; p.q_head=Hq; p.kv_head=Hk; p.q_len=1; p.kv_len=sl; p.head_dim=D;
+        p.use_mask=0; p.is_causal=0; p.causal_offset=0;
+        p.scale=1.0f/sqrtf((float)D);
+        p.q=dQ; p.k=dK; p.v=dV; p.mask=nullptr; p.o=dO;
+
+        DecodeScratch sc;
+        cudaMalloc(&sc.o_part, (size_t)B*Hq*32*D*sizeof(float));
+        cudaMalloc(&sc.ml_part, (size_t)B*Hq*32*2*sizeof(float));
+
+        for (int i=0;i<WARMUP;i++) dispatch_decode(p, sc);
+        cudaDeviceSynchronize();
+        cudaError_t err=cudaGetLastError();
+        if (err!=cudaSuccess){printf("CUDA err: %s\n",cudaGetErrorString(err));return;}
+
+        cudaEvent_t s,e; cudaEventCreate(&s); cudaEventCreate(&e);
+        cudaEventRecord(s);
+        for (int i=0;i<ITERS;i++) dispatch_decode(p, sc);
+        cudaEventRecord(e); cudaEventSynchronize(e);
+        float ms=0; cudaEventElapsedTime(&ms,s,e); ms/=ITERS;
+
+        double flops = 4.0*B*Hq*(double)sl*D;
+        double tflops = flops/(ms*1e-3)/1e12;
+        // HBM traffic: K + V read (B*Hk*sl*D each), bf16; Q/O negligible.
+        double bytes = 2.0 * (2.0*nKV);
+        double gbps = bytes/(ms*1e-3)/1e9;
+
+        char cfg[64];
+        snprintf(cfg, sizeof(cfg),
+                 "B=%2d Hq=%2d Hk=%d q=%4d kv=%4d D=%3d causal=%d",
+                 B,Hq,Hk,1,sl,D,0);
+        printf("%-46s | %7.4f ms | %7.1f GB/s | %6.2f TFLOP/s\n",
+               cfg, ms, gbps, tflops);
+
+        cudaFree(dQ);cudaFree(dK);cudaFree(dV);cudaFree(dO);
+        cudaFree(sc.o_part);cudaFree(sc.ml_part);
+        delete[]tmp; cudaEventDestroy(s); cudaEventDestroy(e);
+    }
+}
+
+int main() {
+    const int configs[][5] = {
+        {1, 2, 1, 64, 32},    // B,Hq,Hk,seq_len,D
+        {1, 32, 4, 512, 128},
+        {1, 32, 4, 1024, 128},
+    };
+    int n_cfgs = sizeof(configs) / sizeof(configs[0]);
+
+    for (int ci = 0; ci < n_cfgs; ci++) {
+        int B = configs[ci][0], Hq = configs[ci][1], Hk = configs[ci][2];
+        int sl = configs[ci][3], D = configs[ci][4], gs = Hq / Hk;
+        printf("=== B=%d Hq=%d Hk=%d seq=%d D=%d gs=%d ===\n", B,Hq,Hk,sl,D,gs);
+
+        size_t nQ = B*Hq*1*D, nKV = B*Hk*sl*D;
+        float *hQ=new float[nQ], *hK=new float[nKV], *hV=new float[nKV];
+        for (size_t i=0;i<nQ;i++) hQ[i]=randf();
+        for (size_t i=0;i<nKV;i++){hK[i]=randf();hV[i]=randf();}
+
+        bool* hMask=new bool[B*sl];
+        for (int i=0;i<B*sl;i++) hMask[i]=true;
+
+        bf16 *dQ,*dK,*dV,*dO,*tmp;
+        bool* dMask;
+        cudaMalloc(&dQ,nQ*2); cudaMalloc(&dK,nKV*2);
+        cudaMalloc(&dV,nKV*2); cudaMalloc(&dO,nQ*2);
+        cudaMalloc(&dMask,B*sl);
+
+        tmp=new bf16[max(nQ,nKV)];
+        for (size_t i=0;i<nQ;i++) tmp[i]=f2bf(hQ[i]);
+        cudaMemcpy(dQ,tmp,nQ*2,cudaMemcpyHostToDevice);
+        for (size_t i=0;i<nKV;i++) tmp[i]=f2bf(hK[i]);
+        cudaMemcpy(dK,tmp,nKV*2,cudaMemcpyHostToDevice);
+        for (size_t i=0;i<nKV;i++) tmp[i]=f2bf(hV[i]);
+        cudaMemcpy(dV,tmp,nKV*2,cudaMemcpyHostToDevice);
+        cudaMemcpy(dMask,hMask,B*sl,cudaMemcpyHostToDevice);
+
+        AttentionParams<bf16> p;
+        p.batch=B; p.q_head=Hq; p.kv_head=Hk; p.q_len=1; p.kv_len=sl; p.head_dim=D;
+        p.use_mask=0; p.is_causal=0; p.causal_offset=0;
+        p.scale=1.0f/sqrtf((float)D);
+        p.q=dQ; p.k=dK; p.v=dV; p.mask=nullptr; p.o=dO;
+
+        // Split-K scratch (max 32 splits), sized for the production MMA path.
+        DecodeScratch sc;
+        cudaMalloc(&sc.o_part, (size_t)B*Hq*32*D*sizeof(float));
+        cudaMalloc(&sc.ml_part, (size_t)B*Hq*32*2*sizeof(float));
+
+        double t0=now_ms();
+        dispatch_decode(p, sc);
+        cudaDeviceSynchronize();
+        double kms=now_ms()-t0;
+        cudaError_t err=cudaGetLastError();
+        if (err!=cudaSuccess){printf("CUDA err: %s\n",cudaGetErrorString(err));return 1;}
+
+        bf16* hOut=new bf16[nQ];
+        cudaMemcpy(hOut,dO,nQ*2,cudaMemcpyDeviceToHost);
+
+        float* ref=new float[nQ];
+        cpu_attention_ref(hQ, hK, hV, hMask, ref, B, Hq, Hk, 1, sl, D, 0, 0);
+
+        float max_err=0;
+        for (size_t i=0;i<nQ;i++){
+            float d=fabsf(bf2f(hOut[i])-ref[i]);
+            if(d>max_err) max_err=d;
+        }
+        printf("kernel: %.3f ms  max_err: %.6e\n\n",kms,max_err);
+
+        cudaFree(dQ);cudaFree(dK);cudaFree(dV);cudaFree(dO);cudaFree(dMask);
+        cudaFree(sc.o_part);cudaFree(sc.ml_part);
+        delete[]hQ;delete[]hK;delete[]hV;delete[]hMask;delete[]hOut;delete[]ref;delete[]tmp;
+    }
+    printf("All tests passed!\n");
+    bench();
+    return 0;
+}

csrc/tests/attn_paged_decode_test.cu

@@ -0,0 +1,346 @@
+// Compile:
+//   nvcc -I csrc -arch=sm_89 -O3 --use_fast_math --ptxas-options=-O3 \
+//        --extra-device-vectorization csrc/tests/attn_paged_decode_test.cu \
+//        -o /tmp/test_paged && /tmp/test_paged
+
+#include <cstring>
+#include "test_utils.cuh"
+#include "../kernels/attn_paged_decode_split_kv.cuh"
+#ifndef ASTRAI_NO_MMA
+#include "../kernels/attn_paged_decode_split_kv_mma.cuh"
+#endif
+
+// Copy contiguous K/V from page pool (reference gather)
+static void gather_kv_cpu(
+    const bf16* h_k_pool, const bf16* h_v_pool,
+    const int64_t* h_pt, int B, int Hkv, int kv_len,
+    int page_size, int head_dim,
+    bf16* h_k, bf16* h_v)
+{
+    int max_pages = (kv_len + page_size - 1) / page_size;
+    size_t page_stride = (size_t)page_size * Hkv * head_dim;
+    for (int b = 0; b < B; b++) {
+        for (int pos = 0; pos < kv_len; pos++) {
+            int log_pg = pos / page_size;
+            int pg_off = pos % page_size;
+            int phys = (int)h_pt[b * max_pages + log_pg];
+            for (int h = 0; h < Hkv; h++) {
+                size_t src_base = (size_t)phys * page_stride
+                                + (size_t)pg_off * Hkv * head_dim
+                                + h * head_dim;
+                size_t dst_base = ((size_t)b * Hkv + h) * kv_len * head_dim + (size_t)pos * head_dim;
+                memcpy(h_k + dst_base, h_k_pool + src_base, head_dim * sizeof(bf16));
+                memcpy(h_v + dst_base, h_v_pool + src_base, head_dim * sizeof(bf16));
+            }
+        }
+    }
+}
+
+template <int HEAD_DIM>
+static void launch_paged_decode(PagedAttentionParams<bf16, float>& p) {
+#ifndef ASTRAI_NO_MMA
+    int G_check = p.q_head / p.kv_head;
+    bool use_mma = !p.use_mask && G_check >= 1 && G_check <= 16 && p.page_size >= 32;
+    if (use_mma) {
+        int tiles_total = (p.kv_len + 32 - 1) / 32;
+        p.num_splits = compute_num_splits(p.batch * p.kv_head, tiles_total);
+        paged_attn_decode_split_kv_mma_kernel<HEAD_DIM, 32>
+            <<<dim3(p.kv_head, p.batch, p.num_splits), 32>>>(p);
+    } else
+#endif
+    {
+        int group_sz = p.q_head / p.kv_head;
+        int chunks_total = (p.kv_len + PDC_CHUNK - 1) / PDC_CHUNK;
+        p.num_splits = compute_num_splits(p.batch * p.kv_head, chunks_total);
+        size_t smem = PDC_CHUNK * p.head_dim * sizeof(bf16);
+        paged_attn_decode_split_kv_kernel<<<
+            dim3(p.batch * p.kv_head, 1, p.num_splits),
+            dim3(32, group_sz), smem>>>(p);
+    }
+    paged_attn_decode_combine_kernel<<<p.batch * p.q_head, p.head_dim>>>(p);
+}
+
+template <int HEAD_DIM>
+static int run_test(int B, int Hq, int Hkv, int kv_len, int page_size, int seed) {
+    printf("B=%d Hq=%d Hkv=%d kv_len=%d page_sz=%d head_dim=%d ... ", B, Hq, Hkv, kv_len, page_size, HEAD_DIM);
+    fflush(stdout);
+
+    int max_pages = (kv_len + page_size - 1) / page_size;
+    int n_phys_pages = B * max_pages;
+
+    size_t sz_q  = (size_t)B * Hq * 1 * HEAD_DIM * sizeof(bf16);
+    size_t sz_o  = sz_q;
+    size_t sz_kv = (size_t)n_phys_pages * page_size * Hkv * HEAD_DIM * sizeof(bf16);
+    size_t sz_pt = (size_t)B * max_pages * sizeof(int64_t);
+    int max_splits = 32;
+    size_t sz_op = (size_t)B * Hq * max_splits * HEAD_DIM * sizeof(float);
+    size_t sz_ml = (size_t)B * Hq * max_splits * 2 * sizeof(float);
+
+    bf16 *d_q, *d_o_paged, *d_o_ref;
+    bf16 *d_k_pool, *d_v_pool;
+    int64_t* d_pt;
+    float *d_op, *d_ml;
+
+    cudaMalloc(&d_q, sz_q);
+    cudaMalloc(&d_o_paged, sz_o);
+    cudaMalloc(&d_o_ref, sz_o);
+    cudaMalloc(&d_k_pool, sz_kv);
+    cudaMalloc(&d_v_pool, sz_kv);
+    cudaMalloc(&d_pt, sz_pt);
+    cudaMalloc(&d_op, sz_op);
+    cudaMalloc(&d_ml, sz_ml);
+
+    srand(seed);
+    auto rnd = [&]() { return (rand() / (float)RAND_MAX) * 2.0f - 1.0f; };
+
+    bf16* h_q = (bf16*)malloc(sz_q);
+    for (int i = 0; i < B * Hq * HEAD_DIM; i++)
+        h_q[i] = __float2bfloat16(rnd());
+    cudaMemcpy(d_q, h_q, sz_q, cudaMemcpyHostToDevice);
+
+    bf16* h_k_pool = (bf16*)malloc(sz_kv);
+    bf16* h_v_pool = (bf16*)malloc(sz_kv);
+    size_t ps = (size_t)page_size * Hkv * HEAD_DIM;
+    for (int pg = 0; pg < n_phys_pages; pg++) {
+        for (int off = 0; off < page_size; off++) {
+            for (int h = 0; h < Hkv; h++) {
+                for (int d = 0; d < HEAD_DIM; d++) {
+                    float v = sinf((float)(pg * 7919 + off * 1049 + h * 331 + d));
+                    size_t idx = (size_t)pg * ps + (size_t)off * Hkv * HEAD_DIM + h * HEAD_DIM + d;
+                    h_k_pool[idx] = __float2bfloat16(v);
+                    h_v_pool[idx] = __float2bfloat16(v * 0.3f);
+                }
+            }
+        }
+    }
+    cudaMemcpy(d_k_pool, h_k_pool, sz_kv, cudaMemcpyHostToDevice);
+    cudaMemcpy(d_v_pool, h_v_pool, sz_kv, cudaMemcpyHostToDevice);
+
+    int64_t* h_pt = (int64_t*)malloc(sz_pt);
+    int next_pg = 0;
+    for (int b = 0; b < B; b++)
+        for (int p = 0; p < max_pages; p++)
+            h_pt[b * max_pages + p] = next_pg++;
+    cudaMemcpy(d_pt, h_pt, sz_pt, cudaMemcpyHostToDevice);
+
+    bf16* h_k_cont = (bf16*)malloc((size_t)B * kv_len * Hkv * HEAD_DIM * sizeof(bf16));
+    bf16* h_v_cont = (bf16*)malloc((size_t)B * kv_len * Hkv * HEAD_DIM * sizeof(bf16));
+    gather_kv_cpu(h_k_pool, h_v_pool, h_pt, B, Hkv, kv_len, page_size, HEAD_DIM, h_k_cont, h_v_cont);
+
+    float* h_q_f = (float*)malloc((size_t)B * Hq * HEAD_DIM * sizeof(float));
+    float* h_k_f = (float*)malloc((size_t)B * kv_len * Hkv * HEAD_DIM * sizeof(float));
+    float* h_v_f = (float*)malloc((size_t)B * kv_len * Hkv * HEAD_DIM * sizeof(float));
+    for (int i = 0; i < B * Hq * HEAD_DIM; i++) h_q_f[i] = bf2f(h_q[i]);
+    for (int i = 0; i < B * kv_len * Hkv * HEAD_DIM; i++) {
+        h_k_f[i] = bf2f(h_k_cont[i]);
+        h_v_f[i] = bf2f(h_v_cont[i]);
+    }
+
+    float* h_o_ref = (float*)calloc(B * Hq * HEAD_DIM, sizeof(float));
+    cpu_attention_ref(h_q_f, h_k_f, h_v_f, nullptr, h_o_ref, B, Hq, Hkv, 1, kv_len, HEAD_DIM, 0, 0);
+
+    float scale_val = 1.0f / sqrtf((float)HEAD_DIM);
+    PagedAttentionParams<bf16, float> p;
+    p.batch = B; p.q_head = Hq; p.kv_head = Hkv; p.q_len = 1;
+    p.kv_len = kv_len; p.head_dim = HEAD_DIM;
+    p.use_mask = 0; p.is_causal = 0; p.causal_offset = 0;
+    p.num_splits = 1; p.scale = scale_val;
+    p.page_size = page_size; p.max_pages = max_pages;
+    p.page_table = d_pt;
+    p.k_cache = d_k_pool; p.v_cache = d_v_pool;
+    p.q = d_q; p.mask = nullptr; p.o = d_o_paged;
+    p.o_part = d_op; p.ml_part = d_ml;
+
+    launch_paged_decode<HEAD_DIM>(p);
+    cudaDeviceSynchronize();
+
+    bf16* h_o_bf16 = (bf16*)malloc(sz_o);
+    cudaMemcpy(h_o_bf16, d_o_paged, sz_o, cudaMemcpyDeviceToHost);
+    float* h_o_paged = (float*)malloc(B * Hq * HEAD_DIM * sizeof(float));
+    for (int i = 0; i < B * Hq * HEAD_DIM; i++)
+        h_o_paged[i] = __bfloat162float(h_o_bf16[i]);
+
+    float max_err = 0.0f;
+    int bad_idx = -1;
+    for (int i = 0; i < B * Hq * HEAD_DIM; i++) {
+        float e = fabsf(h_o_paged[i] - h_o_ref[i]);
+        if (e > max_err) { max_err = e; bad_idx = i; }
+    }
+
+    bool pass = max_err < 0.02f;
+
+    if (pass) {
+        printf("PASS (max_abs_err=%.4e)\n", max_err);
+    } else {
+        int b = bad_idx / (Hq * HEAD_DIM);
+        int h = (bad_idx / HEAD_DIM) % Hq;
+        int d = bad_idx % HEAD_DIM;
+        printf("FAIL (max_abs_err=%.4e at [%d,%d,%d]: ref=%.4f got=%.4f)\n",
+               max_err, b, h, d, h_o_ref[bad_idx], h_o_paged[bad_idx]);
+        printf("  ref[0..7]:");
+        for (int i = 0; i < 8 && i < HEAD_DIM; i++)
+            printf(" %.4f", h_o_ref[i]);
+        printf("\n  got[0..7]:");
+        for (int i = 0; i < 8 && i < HEAD_DIM; i++)
+            printf(" %.4f", h_o_paged[i]);
+        printf("\n");
+    }
+
+    free(h_q); free(h_k_pool); free(h_v_pool); free(h_pt);
+    free(h_k_cont); free(h_v_cont);
+    free(h_q_f); free(h_k_f); free(h_v_f);
+    free(h_o_ref); free(h_o_bf16); free(h_o_paged);
+    cudaFree(d_q); cudaFree(d_o_paged); cudaFree(d_o_ref);
+    cudaFree(d_k_pool); cudaFree(d_v_pool); cudaFree(d_pt);
+    cudaFree(d_op); cudaFree(d_ml);
+
+    return pass ? 0 : 1;
+}
+
+struct TestCase {
+    int head_dim;
+    int B, Hq, Hkv, kv_len, page_size, seed;
+};
+
+static const TestCase TESTS[] = {
+    {128, 1, 1, 1, 8, 128, 1},
+    {128, 1, 4, 4, 128, 128, 2},
+    {128, 2, 4, 4, 256, 128, 3},
+    {128, 1, 4, 1, 64, 64, 4},
+    {128, 1, 8, 2, 64, 128, 5},
+    {128, 2, 16, 4, 128, 128, 6},
+    {64, 1, 4, 2, 32, 128, 7},
+    {256, 1, 2, 1, 16, 128, 8},
+    {32, 1, 4, 2, 32, 64, 9},
+    {128, 3, 8, 2, 256, 128, 10},
+    {128, 2, 32, 8, 512, 128, 11},
+#ifndef ASTRAI_NO_MMA
+    {128, 1, 16, 2, 256, 128, 12},
+    {128, 2, 32, 4, 512, 128, 13},
+#endif
+};
+
+static int dispatch_test(const TestCase& tc) {
+    switch (tc.head_dim) {
+        case 32:  return run_test<32>(tc.B, tc.Hq, tc.Hkv, tc.kv_len, tc.page_size, tc.seed);
+        case 64:  return run_test<64>(tc.B, tc.Hq, tc.Hkv, tc.kv_len, tc.page_size, tc.seed);
+        case 128: return run_test<128>(tc.B, tc.Hq, tc.Hkv, tc.kv_len, tc.page_size, tc.seed);
+        case 256: return run_test<256>(tc.B, tc.Hq, tc.Hkv, tc.kv_len, tc.page_size, tc.seed);
+        default:  return 1;
+    }
+}
+
+// Warmed-up, CUDA-event timed sweep over paged decode configs.
+// Reports per-call latency and effective K/V read bandwidth.
+// Bytes = K + V read through page table (B*Hk*kv*D each), bf16.
+template <int HEAD_DIM>
+static void bench_config(int B, int Hq, int Hkv, int kv_len, int page_size) {
+    int max_pages = (kv_len + page_size - 1) / page_size;
+    int n_phys_pages = B * max_pages;
+
+    size_t sz_q  = (size_t)B * Hq * 1 * HEAD_DIM * sizeof(bf16);
+    size_t sz_kv = (size_t)n_phys_pages * page_size * Hkv * HEAD_DIM * sizeof(bf16);
+    size_t sz_pt = (size_t)B * max_pages * sizeof(int64_t);
+    int max_splits = 32;
+    size_t sz_op = (size_t)B * Hq * max_splits * HEAD_DIM * sizeof(float);
+    size_t sz_ml = (size_t)B * Hq * max_splits * 2 * sizeof(float);
+
+    bf16 *d_q, *d_o, *d_k_pool, *d_v_pool;
+    int64_t* d_pt;
+    float *d_op, *d_ml;
+    cudaMalloc(&d_q, sz_q); cudaMalloc(&d_o, sz_q);
+    cudaMalloc(&d_k_pool, sz_kv); cudaMalloc(&d_v_pool, sz_kv);
+    cudaMalloc(&d_pt, sz_pt);
+    cudaMalloc(&d_op, sz_op); cudaMalloc(&d_ml, sz_ml);
+
+    bf16* tmp = (bf16*)malloc(sz_kv > sz_q ? sz_kv : sz_q);
+    for (size_t i = 0; i < sz_q / sizeof(bf16); i++) tmp[i] = f2bf(randf());
+    cudaMemcpy(d_q, tmp, sz_q, cudaMemcpyHostToDevice);
+    for (size_t i = 0; i < sz_kv / sizeof(bf16); i++) tmp[i] = f2bf(randf());
+    cudaMemcpy(d_k_pool, tmp, sz_kv, cudaMemcpyHostToDevice);
+    cudaMemcpy(d_v_pool, tmp, sz_kv, cudaMemcpyHostToDevice);
+
+    int64_t* h_pt = (int64_t*)malloc(sz_pt);
+    int next_pg = 0;
+    for (int b = 0; b < B; b++)
+        for (int p = 0; p < max_pages; p++)
+            h_pt[b * max_pages + p] = next_pg++;
+    cudaMemcpy(d_pt, h_pt, sz_pt, cudaMemcpyHostToDevice);
+    free(h_pt);
+
+    float scale_val = 1.0f / sqrtf((float)HEAD_DIM);
+    PagedAttentionParams<bf16, float> pa;
+    pa.batch = B; pa.q_head = Hq; pa.kv_head = Hkv; pa.q_len = 1;
+    pa.kv_len = kv_len; pa.head_dim = HEAD_DIM;
+    pa.use_mask = 0; pa.is_causal = 0; pa.causal_offset = 0;
+    pa.num_splits = 1; pa.scale = scale_val;
+    pa.page_size = page_size; pa.max_pages = max_pages;
+    pa.page_table = d_pt;
+    pa.k_cache = d_k_pool; pa.v_cache = d_v_pool;
+    pa.q = d_q; pa.mask = nullptr; pa.o = d_o;
+    pa.o_part = d_op; pa.ml_part = d_ml;
+
+    const int WARMUP = 10, ITERS = 100;
+    for (int i = 0; i < WARMUP; i++) launch_paged_decode<HEAD_DIM>(pa);
+    cudaDeviceSynchronize();
+    CUDA_CHECK(cudaGetLastError());
+
+    cudaEvent_t s, e;
+    cudaEventCreate(&s); cudaEventCreate(&e);
+    cudaEventRecord(s);
+    for (int i = 0; i < ITERS; i++) launch_paged_decode<HEAD_DIM>(pa);
+    cudaEventRecord(e); cudaEventSynchronize(e);
+    float ms = 0; cudaEventElapsedTime(&ms, s, e); ms /= ITERS;
+
+    double flops = 4.0 * B * Hq * (double)kv_len * HEAD_DIM;
+    double tflops = flops / (ms * 1e-3) / 1e12;
+    size_t nKV = (size_t)B * Hkv * kv_len * HEAD_DIM;
+    double bytes = 2.0 * (2.0 * nKV);
+    double gbps = bytes / (ms * 1e-3) / 1e9;
+
+    char cfg[64];
+    snprintf(cfg, sizeof(cfg),
+             "B=%2d Hq=%2d Hk=%d q=%4d kv=%4d D=%3d page=%3d",
+             B, Hq, Hkv, 1, kv_len, HEAD_DIM, page_size);
+    printf("%-46s | %7.4f ms | %7.1f GB/s | %6.2f TFLOP/s\n",
+           cfg, ms, gbps, tflops);
+
+    free(tmp);
+    cudaFree(d_q); cudaFree(d_o);
+    cudaFree(d_k_pool); cudaFree(d_v_pool); cudaFree(d_pt);
+    cudaFree(d_op); cudaFree(d_ml);
+    cudaEventDestroy(s); cudaEventDestroy(e);
+}
+
+static void bench() {
+    printf("\n===== PAGED DECODE BENCH =====\n");
+    printf("%-46s | %10s | %10s | %10s\n",
+           "config", "latency", "bandwidth", "throughput");
+    printf("---------------------------------------------------------------"
+           "----------------------------\n");
+    bench_config<128>(1, 32, 4, 512, 128);
+    bench_config<128>(1, 32, 4, 1024, 128);
+    bench_config<128>(1, 32, 4, 2048, 128);
+    bench_config<128>(1, 32, 4, 4096, 128);
+    bench_config<128>(16, 32, 4, 2048, 128);
+    bench_config<128>(32, 32, 4, 1024, 128);
+}
+
+int main() {
+    int n = sizeof(TESTS) / sizeof(TESTS[0]);
+    int fail = 0;
+    printf("=== Paged Decode vs CPU reference (%d cases) ===\n\n", n);
+
+    for (int i = 0; i < n; i++) {
+        fail += dispatch_test(TESTS[i]);
+        if (fail) break;
+    }
+
+    if (fail) {
+        printf("\nFAILED (%d/%d tests failed)\n", fail, n);
+        return fail;
+    }
+    printf("\nAll %d tests passed!\n", n);
+    bench();
+    return 0;
+}

csrc/tests/attn_prefill_test.cu

@@ -0,0 +1,178 @@
+/*
+Pure-C test:
+nvcc -I csrc -arch=sm_89 -O3 \
+    --use_fast_math --ptxas-options=-O3 --extra-device-vectorization \
+    csrc/tests/attn_prefill_test.cu -o test && ./test
+*/
+
+#include "test_utils.cuh"
+#include "../kernels/attn_prefill_split_q.cuh"
+#ifndef ASTRAI_NO_MMA
+#include "../kernels/attn_prefill_split_q_mma.cuh"
+#endif
+
+// Launch the production prefill path (tensor-core MMA on sm_80+, else the
+// scalar fallback), mirroring dispatch_prefill() in attn_prefill.cu.
+template <int HEAD_DIM>
+static void launch_prefill(AttentionParams<bf16>& p) {
+#ifndef ASTRAI_NO_MMA
+    constexpr int WARPS = 4, BR = 16;
+    constexpr int BC = (HEAD_DIM <= 128) ? 32 : 16;
+    constexpr int MIN_BLOCKS = (HEAD_DIM <= 32) ? 6 : (HEAD_DIM <= 64) ? 4
+                             : (HEAD_DIM <= 128) ? 3 : 2;
+    dim3 grid((p.q_len + BR * WARPS - 1) / (BR * WARPS), p.q_head, p.batch);
+    dim3 block(WARPS * 32, 1, 1);
+    attn_prefill_split_q_mma_kernel<HEAD_DIM, WARPS, BC, MIN_BLOCKS><<<grid, block>>>(p);
+#else
+    constexpr int G = 8, ROWS = 32, P_BC = 32;
+    dim3 grid((p.q_len + ROWS - 1) / ROWS, p.q_head, p.batch);
+    dim3 block(G, ROWS, 1);
+    attn_prefill_split_q_kernel_t<HEAD_DIM, G, ROWS, P_BC><<<grid, block>>>(p);
+#endif
+}
+
+static void dispatch_prefill(AttentionParams<bf16>& p) {
+    switch (p.head_dim) {
+        case 64:  launch_prefill<64>(p);  break;
+        case 128: launch_prefill<128>(p); break;
+        default:  printf("bench: unsupported D=%d\n", p.head_dim);
+    }
+}
+
+// Warmed-up, CUDA-event timed throughput sweep over the production MMA path.
+// Reports per-call latency and effective tensor-core TFLOP/s (2 matmuls:
+// QK^T and P@V, each 2*B*Hq*ql*kl*D flops; halved for causal).
+static void bench() {
+    const int cfgs[][7] = {
+        {1,32,4,512,512,128,0},
+        {1,32,4,1024,1024,128,0},
+        {1,32,4,2048,2048,128,0},
+        {1,32,4,2048,2048,128,1},
+        {4,32,4,2048,2048,128,1},
+        {1,32,4,4096,4096,128,1},
+    };
+    int n = sizeof(cfgs)/sizeof(cfgs[0]);
+    const int WARMUP = 10, ITERS = 50;
+    printf("\n===== PREFILL BENCH (warmup=%d iters=%d) =====\n", WARMUP, ITERS);
+    printf("%-46s | %10s | %10s | %10s\n",
+           "config", "latency", "bandwidth", "throughput");
+    printf("---------------------------------------------------------------"
+           "----------------------------\n");
+
+    for (int ci = 0; ci < n; ci++) {
+        int B=cfgs[ci][0], Hq=cfgs[ci][1], Hk=cfgs[ci][2];
+        int ql=cfgs[ci][3], kl=cfgs[ci][4], D=cfgs[ci][5], causal=cfgs[ci][6];
+        size_t nQ=(size_t)B*Hq*ql*D, nKV=(size_t)B*Hk*kl*D;
+
+        bf16 *dQ,*dK,*dV,*dO,*tmp;
+        cudaMalloc(&dQ,nQ*2); cudaMalloc(&dK,nKV*2);
+        cudaMalloc(&dV,nKV*2); cudaMalloc(&dO,nQ*2);
+        size_t big = nQ>nKV?nQ:nKV; tmp=new bf16[big];
+        for (size_t i=0;i<nQ;i++)  tmp[i]=f2bf(randf());
+        cudaMemcpy(dQ,tmp,nQ*2,cudaMemcpyHostToDevice);
+        for (size_t i=0;i<nKV;i++) tmp[i]=f2bf(randf());
+        cudaMemcpy(dK,tmp,nKV*2,cudaMemcpyHostToDevice);
+        for (size_t i=0;i<nKV;i++) tmp[i]=f2bf(randf());
+        cudaMemcpy(dV,tmp,nKV*2,cudaMemcpyHostToDevice);
+
+        AttentionParams<bf16> p;
+        p.batch=B; p.q_head=Hq; p.kv_head=Hk; p.q_len=ql; p.kv_len=kl; p.head_dim=D;
+        p.use_mask=0; p.is_causal=causal; p.causal_offset=0;
+        p.scale=1.0f/sqrtf((float)D);
+        p.q=dQ; p.k=dK; p.v=dV; p.mask=nullptr; p.o=dO;
+
+        for (int i=0;i<WARMUP;i++) dispatch_prefill(p);
+        cudaDeviceSynchronize();
+        cudaError_t err=cudaGetLastError();
+        if (err!=cudaSuccess){printf("CUDA err: %s\n",cudaGetErrorString(err));return;}
+
+        cudaEvent_t s,e; cudaEventCreate(&s); cudaEventCreate(&e);
+        cudaEventRecord(s);
+        for (int i=0;i<ITERS;i++) dispatch_prefill(p);
+        cudaEventRecord(e); cudaEventSynchronize(e);
+        float ms=0; cudaEventElapsedTime(&ms,s,e); ms/=ITERS;
+
+        double flops = 4.0*B*Hq*(double)ql*kl*D;
+        if (causal) flops *= 0.5;
+        double tflops = flops/(ms*1e-3)/1e12;
+        // HBM traffic: Q + O (B*Hq*ql*D each) + K + V (B*Hk*kl*D each), bf16.
+        double bytes = 2.0 * (2.0*nQ + 2.0*nKV);
+        double gbps = bytes/(ms*1e-3)/1e9;
+
+        char cfg[64];
+        snprintf(cfg, sizeof(cfg),
+                 "B=%2d Hq=%2d Hk=%d q=%4d kv=%4d D=%3d causal=%d",
+                 B,Hq,Hk,ql,kl,D,causal);
+        printf("%-46s | %7.4f ms | %7.1f GB/s | %6.2f TFLOP/s\n",
+               cfg, ms, gbps, tflops);
+
+        cudaFree(dQ);cudaFree(dK);cudaFree(dV);cudaFree(dO);
+        delete[]tmp; cudaEventDestroy(s); cudaEventDestroy(e);
+    }
+}
+
+int main() {
+    const int configs[][7] = {
+        {1,2,1,64,128,64,0},     // tiny: B,Hq,Hk,q,kv,D,causal
+        {1,32,4,512,512,128,0},  // standard
+        {1,32,4,128,256,128,0},  // medium
+        {1,4,2,256,256,128,1},   // causal
+    };
+    int n_configs = sizeof(configs) / sizeof(configs[0]);
+
+    for (int ci = 0; ci < n_configs; ci++) {
+        int B=configs[ci][0], Hq=configs[ci][1], Hk=configs[ci][2];
+        int ql=configs[ci][3], kl=configs[ci][4], D=configs[ci][5];
+        int causal=configs[ci][6];
+        printf("=== B=%d Hq=%d Hk=%d q=%d kv=%d D=%d causal=%d ===\n",
+               B,Hq,Hk,ql,kl,D,causal);
+
+        size_t nQ = B*Hq*ql*D, nKV = B*Hk*kl*D;
+        float *hQ=new float[nQ], *hK=new float[nKV], *hV=new float[nKV];
+        for (size_t i=0;i<nQ;i++) hQ[i]=randf();
+        for (size_t i=0;i<nKV;i++){hK[i]=randf();hV[i]=randf();}
+
+        bf16 *dQ,*dK,*dV,*dO,*tmp;
+        cudaMalloc(&dQ,nQ*2); cudaMalloc(&dK,nKV*2);
+        cudaMalloc(&dV,nKV*2); cudaMalloc(&dO,nQ*2);
+        tmp=new bf16[max(nQ,nKV)];
+        for (size_t i=0;i<nQ;i++) tmp[i]=f2bf(hQ[i]);
+        cudaMemcpy(dQ,tmp,nQ*2,cudaMemcpyHostToDevice);
+        for (size_t i=0;i<nKV;i++) tmp[i]=f2bf(hK[i]);
+        cudaMemcpy(dK,tmp,nKV*2,cudaMemcpyHostToDevice);
+        for (size_t i=0;i<nKV;i++) tmp[i]=f2bf(hV[i]);
+        cudaMemcpy(dV,tmp,nKV*2,cudaMemcpyHostToDevice);
+
+        AttentionParams<bf16> p;
+        p.batch=B; p.q_head=Hq; p.kv_head=Hk; p.q_len=ql; p.kv_len=kl; p.head_dim=D;
+        p.use_mask=0; p.is_causal=causal; p.causal_offset=0;
+        p.scale=1.0f/sqrtf((float)D);
+        p.q=dQ; p.k=dK; p.v=dV; p.mask=nullptr; p.o=dO;
+
+        double t0=now_ms();
+        dispatch_prefill(p);
+        cudaDeviceSynchronize();
+        double kms=now_ms()-t0;
+        cudaError_t err=cudaGetLastError();
+        if (err!=cudaSuccess){printf("CUDA err: %s\n",cudaGetErrorString(err));return 1;}
+
+        bf16* hOut=new bf16[nQ];
+        cudaMemcpy(hOut,dO,nQ*2,cudaMemcpyDeviceToHost);
+
+        float* ref=new float[nQ];
+        cpu_attention_ref(hQ, hK, hV, nullptr, ref, B, Hq, Hk, ql, kl, D, causal, 0);
+
+        float max_err=0;
+        for (size_t i=0;i<nQ;i++) {
+            float d=fabsf(bf2f(hOut[i])-ref[i]);
+            if(d>max_err) max_err=d;
+        }
+        printf("kernel: %.3f ms  max_err: %.6e\n\n",kms,max_err);
+
+        cudaFree(dQ);cudaFree(dK);cudaFree(dV);cudaFree(dO);
+        delete[]hQ;delete[]hK;delete[]hV;delete[]hOut;delete[]ref;delete[]tmp;
+    }
+    printf("All tests passed!\n");
+    bench();
+    return 0;
+}

csrc/tests/test_utils.cuh

@@ -0,0 +1,90 @@
+#pragma once
+
+#include <cstdio>
+#include <cstdlib>
+#include <cmath>
+#include <chrono>
+#include <cuda_bf16.h>
+
+using bf16 = __nv_bfloat16;
+
+inline bf16 f2bf(float x) { return __float2bfloat16(x); }
+inline float bf2f(bf16 x) { return __bfloat162float(x); }
+
+inline float randf() { return (float)rand() / (float)RAND_MAX - 0.5f; }
+
+inline double now_ms() {
+    using namespace std::chrono;
+    return duration_cast<milliseconds>(steady_clock::now().time_since_epoch()).count();
+}
+
+inline int compute_num_splits(int base_blocks, int tiles_total) {
+    int sm_count = 0;
+    cudaDeviceGetAttribute(&sm_count, cudaDevAttrMultiProcessorCount, 0);
+    int n = (2 * sm_count + base_blocks - 1) / base_blocks;
+    if (n > tiles_total) n = tiles_total;
+    if (n > 32) n = 32;
+    if (n < 1) n = 1;
+    return n;
+}
+
+#define CUDA_CHECK(call) \
+    do { \
+        cudaError_t _e = (call); \
+        if (_e != cudaSuccess) { \
+            printf("CUDA error %s at %s:%d\n", cudaGetErrorString(_e), __FILE__, __LINE__); \
+            exit(1); \
+        } \
+    } while (0)
+
+// Generic CPU reference for multi-query / grouped-query attention.
+// Tensor shapes (all float*):
+//   Q : [B, Hq, q_len, D]
+//   K : [B, Hk, kv_len, D]
+//   V : [B, Hk, kv_len, D]
+//   O : [B, Hq, q_len, D]
+// mask: if q_len == 1, shape is [B, kv_len]; otherwise mask is not supported.
+static void cpu_attention_ref(
+    const float* Q, const float* K, const float* V, const bool* mask,
+    float* O, int B, int Hq, int Hk, int q_len, int kv_len, int D,
+    int is_causal, int causal_offset
+) {
+    float scale = 1.0f / sqrtf((float)D);
+    int n_rep = Hq / Hk;
+    for (int b = 0; b < B; b++) {
+        for (int h = 0; h < Hq; h++) {
+            int kv_h = h / n_rep;
+            for (int qi = 0; qi < q_len; qi++) {
+                float mv = -INFINITY, sv = 0.0f;
+                float accum[256] = {0.0f};
+                int lim = kv_len;
+                if (is_causal) {
+                    int c = qi + causal_offset + 1;
+                    lim = (c < kv_len) ? c : kv_len;
+                }
+                for (int kj = 0; kj < lim; kj++) {
+                    if (mask != nullptr && q_len == 1) {
+                        if (!mask[b * kv_len + kj]) continue;
+                    }
+                    float dot = 0.0f;
+                    size_t q_idx = ((size_t)b * Hq + h) * q_len + qi;
+                    size_t kv_idx = ((size_t)b * Hk + kv_h) * kv_len + kj;
+                    for (int d = 0; d < D; d++)
+                        dot += Q[q_idx * D + d] * K[kv_idx * D + d];
+                    dot *= scale;
+                    float nm = fmaxf(mv, dot);
+                    float a = expf(mv - nm);
+                    float b_exp = expf(dot - nm);
+                    sv = sv * a + b_exp;
+                    for (int d = 0; d < D; d++)
+                        accum[d] = accum[d] * a + V[kv_idx * D + d] * b_exp;
+                    mv = nm;
+                }
+                float inv = 1.0f / sv;
+                size_t o_idx = ((size_t)b * Hq + h) * q_len + qi;
+                for (int d = 0; d < D; d++)
+                    O[o_idx * D + d] = accum[d] * inv;
+            }
+        }
+    }
+}

pyproject.toml

@@ -9,8 +9,8 @@ readme = "README.md"
 requires-python = ">=3.12"
 dependencies = [
     "h5py==3.15.1",
-    "numpy==2.3.2",
-    "torch==2.7.1",
+    "numpy==2.4.4",
+    "torch==2.11.0",
     "tokenizers==0.21.4",
     "tqdm==4.67.1",
     "safetensors==0.5.3",
@@ -37,7 +37,7 @@ dev = ["pytest==9.0.2", "ruff"]
 where = ["."]
 
 [tool.pip]
-extra-index-url = "https://download.pytorch.org/whl/cu126"
+extra-index-url = "https://download.pytorch.org/whl/cu128"
 
 [tool.setuptools.dynamic]
 version = { attr = "astrai.__version__" }

scripts/demo/stream_chat.py

@@ -1,3 +1,4 @@
+from argparse import ArgumentParser
 from pathlib import Path
 
 import torch
@@ -7,42 +8,82 @@ from astrai.model import AutoModel
 from astrai.tokenize import AutoTokenizer
 
 PROJECT_ROOT = Path(__file__).resolve().parents[2]
-PARAMETER_ROOT = Path(PROJECT_ROOT, "params")
+
+
+def parse_args():
+    parser = ArgumentParser(description="Interactive streaming chat")
+    parser.add_argument(
+        "--model_path",
+        type=Path,
+        default=PROJECT_ROOT / "params",
+        help="Path to model weights (params/ or checkpoint/epoch_N_step_M/)",
+    )
+    parser.add_argument(
+        "--temperature",
+        type=float,
+        default=0.8,
+        help="Sampling temperature (default: 0.8)",
+    )
+    parser.add_argument(
+        "--top_p",
+        type=float,
+        default=0.95,
+        help="Top-p sampling threshold",
+    )
+    parser.add_argument(
+        "--top_k",
+        type=int,
+        default=50,
+        help="Top-k sampling threshold",
+    )
+    parser.add_argument(
+        "--max_tokens",
+        type=int,
+        default=2048,
+        help="Maximum tokens to generate",
+    )
+    parser.add_argument(
+        "--system_prompt",
+        type=str,
+        default="You are a helpful assistant.",
+        help="Optional system prompt",
+    )
+    return parser.parse_args()
 
 
 def chat():
-    model = AutoModel.from_pretrained(PARAMETER_ROOT)
-    tokenizer = AutoTokenizer.from_pretrained(PARAMETER_ROOT)
-    model.to(device="cuda", dtype=torch.bfloat16)
+    args = parse_args()
+    model_path = args.model_path
 
-    messages = [{"role": "system", "content": "You are a helpful assistant."}]
+    model = AutoModel.from_pretrained(model_path)
+    tokenizer = AutoTokenizer.from_pretrained(model_path)
+    model.to(device="cuda", dtype=torch.bfloat16)
     engine = InferenceEngine(model=model, tokenizer=tokenizer)
 
+    messages = [{"role": "system", "content": args.system_prompt}]
+
     while True:
         query = input(">> ")
         if query == "!exit":
             break
 
-        # Add user message
         messages.append({"role": "user", "content": query})
 
-        # Generate response
         full_response = ""
         prompt = tokenizer.apply_chat_template(messages, tokenize=False)
 
         for token in engine.generate(
             prompt=prompt,
             stream=True,
-            max_tokens=2048,
-            temperature=0.8,
-            top_p=0.95,
-            top_k=50,
+            max_tokens=args.max_tokens,
+            temperature=args.temperature,
+            top_p=args.top_p,
+            top_k=args.top_k,
         ):
             print(token, end="", flush=True)
             full_response += token
 
         print()
-        # Add assistant response to messages
         messages.append({"role": "assistant", "content": full_response.strip()})
 
 

scripts/eval/analyze_weights.py

@@ -0,0 +1,307 @@
+"""SVD effective rank & weight statistics analysis for model checkpoints."""
+
+import argparse
+import json
+from pathlib import Path
+
+import safetensors.torch
+import torch
+
+
+def effective_rank_metrics(w: torch.Tensor) -> dict:
+    if w.ndim == 1:
+        return {"shape": tuple(w.shape), "is_1d": True}
+
+    w = w.float()
+    s = torch.linalg.svdvals(w)
+    s_sq = s**2
+    total = s_sq.sum()
+    cumsum = torch.cumsum(s_sq, dim=0) / total
+
+    min_dim = min(w.shape[0], w.shape[1])
+    er_90 = (cumsum < 0.90).sum().item() + 1
+    er_95 = (cumsum < 0.95).sum().item() + 1
+    er_99 = (cumsum < 0.99).sum().item() + 1
+
+    p = s_sq / total
+    p = p[p > 1e-30]
+    entropy = -(p * torch.log(p)).sum()
+    entropic_rank = torch.exp(entropy).item()
+
+    return {
+        "shape": tuple(w.shape),
+        "min_dim": min_dim,
+        "er_90": er_90,
+        "er_95": er_95,
+        "er_99": er_99,
+        "er_99_norm": er_99 / min_dim,
+        "er_95_norm": er_95 / min_dim,
+        "entropic_rank": entropic_rank,
+        "entropic_rank_norm": entropic_rank / min_dim,
+        "top1_ratio": s[0].item() / s.sum().item(),
+        "top5_ratio": s[:5].sum().item() / s.sum().item(),
+        "decay_ratio": s[-1].item() / s[0].item(),
+        "condition_number": s[0].item() / s[-1].item(),
+        "mean": w.mean().item(),
+        "std": w.std().item(),
+        "min": w.min().item(),
+        "max": w.max().item(),
+    }
+
+
+def format_header(headers: list[str], widths: list[int]) -> str:
+    return "".join(h.ljust(w) for h, w in zip(headers, widths))
+
+
+def format_row(values: list[str], widths: list[int]) -> str:
+    return "".join(v.ljust(w) for v, w in zip(values, widths))
+
+
+def group_by_component(results: dict[str, dict]) -> dict[str, list[dict]]:
+    groups: dict[str, list[dict]] = {}
+    for key, r in results.items():
+        parts = key.split(".")
+        if parts[0] == "layers" and len(parts) >= 3:
+            sub = parts[2:]
+            if sub[0] == "attention":
+                comp = f"attn.{sub[1]}"
+            elif sub[0] == "mlp":
+                comp = f"mlp.{sub[1]}"
+            elif sub[0] == "input_norm":
+                comp = "input_norm"
+            elif sub[0] == "post_attention_norm":
+                comp = "post_attn_norm"
+            else:
+                comp = ".".join(sub)
+        else:
+            comp = key
+        groups.setdefault(comp, []).append(r)
+    return groups
+
+
+def print_component_summary(results: dict[str, dict], title: str):
+    groups = group_by_component(results)
+    matrix_groups = {
+        k: [v for v in vs if not v.get("is_1d")]
+        for k, vs in groups.items()
+        if any(not v.get("is_1d") for v in vs)
+    }
+
+    widths = [20, 12, 12, 12, 12, 12]
+    print(f"\n{title}")
+    print(
+        format_header(
+            ["Component", "N", "ER@99%", "EntRank%", "Top1 σ(%)", "Cond. Num"], widths
+        )
+    )
+    print("-" * sum(widths))
+
+    for name in sorted(matrix_groups.keys()):
+        items = matrix_groups[name]
+        n = len(items)
+        print(
+            format_row(
+                [
+                    name,
+                    str(n),
+                    f"{sum(r['er_99_norm'] for r in items) / n:.4f}",
+                    f"{sum(r['entropic_rank_norm'] for r in items) / n:.4f}",
+                    f"{sum(r['top1_ratio'] for r in items) / n:.4f}",
+                    f"{sum(r['condition_number'] for r in items) / n:.1f}",
+                ],
+                widths,
+            )
+        )
+
+    all_er = [
+        r["er_99_norm"]
+        for vs in matrix_groups.values()
+        for r in vs
+        if "_norm" not in r or not r.get("is_1d")
+    ]
+    if all_er:
+        m = sum(all_er) / len(all_er)
+        print(f"\n  Overall Mean ER@99: {m:.4f} ({m * 100:.1f}% of dimension)")
+        if m > 0.85:
+            print("  → HIGH utilization: model near capacity → need more params")
+        elif m > 0.5:
+            print("  → MODERATE utilization: some headroom left")
+        else:
+            print("  → LOW utilization: significant unused capacity")
+
+
+def print_layer_grid(results: dict[str, dict]):
+    comps = [
+        "attn.q_proj",
+        "attn.k_proj",
+        "attn.v_proj",
+        "attn.o_proj",
+        "mlp.up",
+        "mlp.gate",
+        "mlp.down",
+    ]
+    widths = [6] + [10] * len(comps)
+    metric = "er_99_norm"
+
+    print(f"\n--- Per-Layer Effective Rank (99% energy) ---")
+    print(format_header(["Layer"] + comps, widths))
+    print("-" * sum(widths))
+
+    layer_data: dict[int, dict[str, dict]] = {}
+    for key, r in results.items():
+        parts = key.split(".")
+        if parts[0] != "layers":
+            continue
+        li = int(parts[1])
+        sub = parts[2:]
+        if sub[0] == "attention":
+            cname = f"attn.{sub[1]}"
+        elif sub[0] == "mlp":
+            cname = f"mlp.{sub[1]}"
+        else:
+            continue
+        layer_data.setdefault(li, {})[cname] = r
+
+    for li in sorted(layer_data):
+        values = [str(li)]
+        for c in comps:
+            v = layer_data[li].get(c, {}).get(metric, 0)
+            values.append(f"{v:.4f}")
+        print(format_row(values, widths))
+
+
+def print_weight_stats(results: dict[str, dict]):
+    groups = group_by_component(results)
+    widths = [20, 12, 12, 12, 12]
+    print(f"\n--- Weight Value Statistics ---")
+    print(format_header(["Component", "Mean", "Std", "Min", "Max"], widths))
+    print("-" * sum(widths))
+
+    for name in sorted(groups.keys()):
+        items = groups[name]
+        means = [r.get("mean", 0) for r in items]
+        stds = [r.get("std", 0) for r in items]
+        mins = [r.get("min", 0) for r in items]
+        maxs = [r.get("max", 0) for r in items]
+        g_mean = sum(means) / len(means)
+        g_std = sum(stds) / len(stds)
+        g_min = min(mins)
+        g_max = max(maxs)
+        print(
+            format_row(
+                [
+                    name,
+                    f"{g_mean:.6f}",
+                    f"{g_std:.6f}",
+                    f"{g_min:.6f}",
+                    f"{g_max:.6f}",
+                ],
+                widths,
+            )
+        )
+
+
+def print_params_summary(results: dict[str, dict]):
+    total_2d = sum(
+        r["shape"][0] * r["shape"][1] for r in results.values() if not r.get("is_1d")
+    )
+    total_1d = sum(r["shape"][0] for r in results.values() if r.get("is_1d"))
+    print(f"\n  Total 2D params: {total_2d:,}")
+    print(f"  Total 1D params: {total_1d:,}")
+    print(f"  Total params:     {total_2d + total_1d:,}")
+
+
+def main():
+    parser = argparse.ArgumentParser(
+        description="SVD effective rank & weight statistics of a model checkpoint."
+    )
+    parser.add_argument(
+        "--ckpt_dir",
+        type=str,
+        required=True,
+        help="Path to checkpoint directory (containing model.safetensors + config.json).",
+    )
+    parser.add_argument(
+        "--compare",
+        type=str,
+        nargs="*",
+        help="Additional checkpoint directories to compare against.",
+    )
+    parser.add_argument(
+        "--no_svd",
+        action="store_true",
+        help="Skip SVD analysis, only show weight statistics (mean/std/min/max).",
+    )
+    args = parser.parse_args()
+
+    def analyze_one(ckpt_dir: str, label: str):
+        ckpt_dir = Path(ckpt_dir)
+        weights_path = ckpt_dir / "model.safetensors"
+        if not weights_path.exists():
+            print(f"ERROR: {weights_path} not found")
+            return {}
+
+        meta = {}
+        meta_path = ckpt_dir / "meta.json"
+        if meta_path.exists():
+            with open(meta_path) as f:
+                meta = json.load(f)
+
+        print(f"\n{'=' * 70}")
+        print(f"  {label}: {ckpt_dir}")
+        if meta:
+            print(
+                f"  Iteration: {meta.get('iteration', '?')}, "
+                f"Strategy: {meta.get('strategy', '?')}, "
+                f"nprocs={meta.get('nprocs', '?')}"
+            )
+        print(f"{'=' * 70}")
+
+        print(f"Loading weights...")
+        sd = safetensors.torch.load_file(str(weights_path))
+        print(f"  {len(sd)} keys loaded")
+
+        weight_keys = [
+            k
+            for k in sd
+            if ".weight" in k and "rotary_embedding" not in k and "freqs_cis" not in k
+        ]
+
+        results = {}
+        if not args.no_svd:
+            print(f"Computing SVD on {len(weight_keys)} tensors...")
+            for i, k in enumerate(sorted(weight_keys)):
+                print(f"  [{i + 1}/{len(weight_keys)}] {k:<60s}", end="\r")
+                results[k] = effective_rank_metrics(sd[k])
+            print()
+        else:
+            print(f"Computing stats on {len(weight_keys)} tensors (no SVD)...")
+            for i, k in enumerate(sorted(weight_keys)):
+                t = sd[k]
+                results[k] = {
+                    "shape": tuple(t.shape),
+                    "is_1d": t.ndim == 1,
+                    "mean": t.float().mean().item(),
+                    "std": t.float().std().item(),
+                    "min": t.float().min().item(),
+                    "max": t.float().max().item(),
+                }
+
+        print_params_summary(results)
+        if not args.no_svd:
+            print_component_summary(
+                results, "\n=== SVD Effective Rank by Component ==="
+            )
+            print_layer_grid(results)
+        print_weight_stats(results)
+        return results
+
+    analyze_one(args.ckpt_dir, "Primary")
+
+    if args.compare:
+        for cdir in args.compare:
+            analyze_one(cdir, "Compare")
+
+
+if __name__ == "__main__":
+    main()

scripts/eval/evaluate_humaneval.py

@@ -0,0 +1,393 @@
+"""HumanEval benchmark — functional pipeline design.
+
+Pipeline:
+    load -> generate -> extract -> test -> score -> report
+
+Each stage is a pure function (except GPU/CPU-bound I/O stages).
+Config is a single dataclass; side effects are isolated at pipeline boundaries.
+"""
+
+import argparse
+import itertools
+import json
+import os
+import re
+import subprocess
+import sys
+from dataclasses import dataclass
+from math import prod
+from typing import Dict, Iterator, List, Optional, Sequence, Tuple
+
+import numpy as np
+import torch
+import tqdm
+
+from astrai.inference import InferenceEngine
+from astrai.model import AutoModel
+from astrai.tokenize import AutoTokenizer
+
+# ---------------------------------------------------------------------------
+# Config
+# ---------------------------------------------------------------------------
+
+HUMANEVAL_URL = (
+    "https://github.com/openai/human-eval/raw/master/data/HumanEval.jsonl.gz"
+)
+
+STOP_SEQUENCES = [
+    "\nclass ",
+    "\ndef ",
+    "\n# ",
+    "\nif __name__",
+    "\nprint(",
+    "\n\n\n",
+]
+
+
+@dataclass
+class EvalConfig:
+    param_path: str = "./params"
+    data_path: str = "./humaneval/HumanEval.jsonl"
+    output: Optional[str] = None
+
+    test_only: Optional[str] = None
+    generate_only: bool = False
+
+    num_samples: int = 200
+    max_tokens: int = 512
+    temperature: float = 0.8
+    top_p: float = 0.95
+    top_k: int = 50
+    batch_size: int = 32
+    test_timeout: float = 3.0
+    test_workers: int = 8
+    k_values: Tuple[int, ...] = (1, 10, 100)
+    problem_indices: Optional[List[int]] = None
+
+
+def download(url: str, path: str):
+    if os.path.exists(path):
+        return
+    import gzip
+    import urllib.request
+
+    os.makedirs(os.path.dirname(path) or ".", exist_ok=True)
+    print(f"Downloading {url} ...")
+    tmp = path + ".tmp"
+    urllib.request.urlretrieve(url, tmp)
+    with gzip.open(tmp, "rb") as f_in:
+        with open(path, "wb") as f_out:
+            f_out.write(f_in.read())
+    os.remove(tmp)
+    print(f"  saved to {path}")
+
+
+def load_jsonl(path: str) -> List[dict]:
+    rows = []
+    with open(path, encoding="utf-8") as f:
+        for line in f:
+            line = line.strip()
+            if line:
+                rows.append(json.loads(line))
+    return rows
+
+
+def save_json(path: str, data):
+    with open(path, "w", encoding="utf-8") as f:
+        json.dump(data, f, indent=2, ensure_ascii=False)
+
+
+def create_engine(param_path: str, batch_size: int) -> InferenceEngine:
+    model = AutoModel.from_pretrained(param_path)
+    tokenizer = AutoTokenizer.from_pretrained(param_path)
+    model.to(device="cuda", dtype=torch.bfloat16)
+    return InferenceEngine(
+        model=model,
+        tokenizer=tokenizer,
+        max_batch_size=batch_size,
+    )
+
+
+def trim_stop(text: str) -> str:
+    for stop in STOP_SEQUENCES:
+        idx = text.find(stop)
+        if idx != -1:
+            text = text[:idx]
+    return text
+
+
+def extract_body(code: str, entry_point: str) -> Optional[str]:
+    pattern = rf"def\s+{re.escape(entry_point)}\b[^:]*:"
+    match = re.search(pattern, code)
+    if not match:
+        return code
+
+    lines = code[match.end() :].split("\n")
+    body_lines = []
+    started = False
+
+    for line in lines:
+        stripped = line.rstrip()
+        if not stripped and not started:
+            continue
+        if not stripped and started:
+            body_lines.append("")
+            continue
+        if not started:
+            started = True
+        if stripped.lstrip() == stripped and started:
+            break
+        body_lines.append(stripped)
+
+    body = "\n".join(body_lines)
+    return body if body.strip() else None
+
+
+def deduplicate(seq: Sequence[str]) -> List[str]:
+    seen = set()
+    return [x for x in seq if not (x in seen or seen.add(x))]
+
+
+def generate_batch(
+    engine: InferenceEngine,
+    prompt: str,
+    n: int,
+    batch_size: int,
+    max_tokens: int,
+    temperature: float,
+    top_p: float,
+    top_k: int,
+) -> List[str]:
+    completions = []
+    remaining = n
+    while remaining > 0:
+        current = min(batch_size, remaining)
+        outputs = engine.generate(
+            prompt=[prompt] * current,
+            stream=False,
+            max_tokens=max_tokens,
+            temperature=temperature,
+            top_p=top_p,
+            top_k=top_k,
+        )
+        completions.extend(outputs if isinstance(outputs, list) else [outputs])
+        remaining -= current
+    return deduplicate(completions)
+
+
+def extract_completions(
+    raw: Sequence[str],
+    entry_point: str,
+) -> List[str]:
+    bodies = []
+    for r in raw:
+        t = trim_stop(r)
+        body = extract_body(t, entry_point)
+        if body:
+            bodies.append(body)
+    return bodies
+
+
+def generate_all(
+    engine: InferenceEngine,
+    problems: Sequence[dict],
+    cfg: EvalConfig,
+) -> List[dict]:
+    results = []
+    for problem in tqdm.tqdm(problems, desc="Generating", unit="problem"):
+        raw = generate_batch(
+            engine,
+            problem["prompt"],
+            cfg.num_samples,
+            cfg.batch_size,
+            cfg.max_tokens,
+            cfg.temperature,
+            cfg.top_p,
+            cfg.top_k,
+        )
+        bodies = extract_completions(raw, problem["entry_point"])
+        results.append(
+            dict(
+                task_id=problem["task_id"],
+                entry_point=problem["entry_point"],
+                prompt=problem["prompt"],
+                test=problem["test"],
+                completions=bodies,
+            )
+        )
+    return results
+
+
+def execute_one(args: tuple) -> bool:
+    full_code, entry_point, timeout = args
+    try:
+        r = subprocess.run(
+            [sys.executable, "-c", full_code],
+            capture_output=True,
+            timeout=timeout,
+        )
+        return r.returncode == 0
+    except subprocess.TimeoutExpired:
+        return False
+    except Exception:
+        return False
+
+
+def test_one(item: dict, cfg: EvalConfig) -> Tuple[str, int, int]:
+    from concurrent.futures import ProcessPoolExecutor
+
+    task_id = item["task_id"]
+    completions = item["completions"]
+    codes = [
+        (
+            item["prompt"] + c + "\n" + item["test"],
+            item["entry_point"],
+            cfg.test_timeout,
+        )
+        for c in completions
+    ]
+    n = len(codes)
+    passed = 0
+    with ProcessPoolExecutor(max_workers=cfg.test_workers) as pool:
+        for ok in pool.map(execute_one, codes):
+            if ok:
+                passed += 1
+    return task_id, n, passed
+
+
+def test_all(
+    items: Sequence[dict],
+    cfg: EvalConfig,
+) -> Iterator[Tuple[str, int, int]]:
+    for item in tqdm.tqdm(items, desc="Testing", unit="problem"):
+        yield test_one(item, cfg)
+
+
+def pass_at_k(n: int, c: int, k: int) -> float:
+    if n - c < k:
+        return 1.0
+    return 1.0 - float(prod(1.0 - k / np.arange(n - c + 1, n + 1)))
+
+
+def score_results(
+    results: Iterator[Tuple[str, int, int]],
+    k_values: Tuple[int, ...],
+) -> Dict:
+    # filter to k <= n (peek first result to get n)
+    first = next(results)
+    results = itertools.chain([first], results)
+    n = first[1]
+    k_values = tuple(k for k in k_values if k <= n)
+
+    scores = {k: [] for k in k_values}
+    output = {}
+    for task_id, n, passed in results:
+        entry = {"task_id": task_id, "n": n, "passed": passed}
+        for k in k_values:
+            pk = round(pass_at_k(n, passed, k), 4)
+            entry[f"pass@{k}"] = pk
+            scores[k].append(pk)
+        output[task_id] = entry
+
+    summary = {}
+    for k in k_values:
+        vals = scores[k]
+        summary[f"pass@{k}"] = round(float(np.mean(vals)), 4)
+    output["_summary"] = summary
+    return output
+
+
+def run_pipeline(cfg: EvalConfig) -> Dict:
+    if cfg.test_only:
+        with open(cfg.test_only, encoding="utf-8") as f:
+            generated = json.load(f)
+    else:
+        download(HUMANEVAL_URL, cfg.data_path)
+
+        problems = load_jsonl(cfg.data_path)
+        if cfg.problem_indices:
+            problems = [problems[i] for i in cfg.problem_indices if i < len(problems)]
+
+        engine = create_engine(cfg.param_path, cfg.batch_size)
+
+        try:
+            generated = generate_all(engine, problems, cfg)
+        finally:
+            engine.shutdown()
+
+        if cfg.output:
+            mid = cfg.output.replace(".json", "_completions.json")
+            save_json(mid, generated)
+            print(f"Completions saved to {mid}")
+
+        if cfg.generate_only:
+            return {}
+
+    results = test_all(generated, cfg)
+    scored = score_results(results, cfg.k_values)
+    return scored
+
+
+def parse_args(argv: Optional[List[str]] = None) -> EvalConfig:
+    p = argparse.ArgumentParser(description="HumanEval benchmark")
+    p.add_argument("--param_path", type=str, default="./params")
+    p.add_argument("--data_path", type=str, default="./humaneval/HumanEval.jsonl")
+    p.add_argument("--output", type=str, default=None)
+    p.add_argument(
+        "--test_only",
+        type=str,
+        default=None,
+        help="Skip generation, test existing completions JSON",
+    )
+    p.add_argument(
+        "--generate_only", action="store_true", help="Only generate, skip testing"
+    )
+    p.add_argument("--num_samples", type=int, default=200)
+    p.add_argument("--max_tokens", type=int, default=512)
+    p.add_argument("--temperature", type=float, default=0.8)
+    p.add_argument("--top_p", type=float, default=0.95)
+    p.add_argument("--top_k", type=int, default=50)
+    p.add_argument("--batch_size", type=int, default=32)
+    p.add_argument("--test_workers", type=int, default=8)
+    p.add_argument("--test_timeout", type=float, default=3.0)
+    p.add_argument("--problems", type=int, nargs="+", default=None)
+    args = p.parse_args(argv)
+
+    return EvalConfig(
+        param_path=args.param_path,
+        data_path=args.data_path,
+        output=args.output,
+        test_only=args.test_only,
+        generate_only=args.generate_only,
+        num_samples=args.num_samples,
+        max_tokens=args.max_tokens,
+        temperature=args.temperature,
+        top_p=args.top_p,
+        top_k=args.top_k,
+        batch_size=args.batch_size,
+        test_workers=args.test_workers,
+        test_timeout=args.test_timeout,
+        problem_indices=args.problems,
+    )
+
+
+def report(scored: Dict):
+    summary = scored.pop("_summary", {})
+    print(f"\n{'=' * 60}")
+    for k, v in summary.items():
+        print(f"  {k}: {v:.2%}")
+    print(f"{'=' * 60}")
+    scored["_summary"] = summary
+
+
+def main():
+    cfg = parse_args()
+    scored = run_pipeline(cfg)
+    report(scored)
+    if cfg.output:
+        save_json(cfg.output, scored)
+        print(f"Results saved to {cfg.output}")
+
+
+if __name__ == "__main__":
+    main()

scripts/eval/evaluate_ifd.py

@@ -0,0 +1,474 @@
+"""IFD (Instruction Following Difficulty) data quality scoring.
+
+IFD = conditional_NLL / unconditional_NLL
+
+- Messages format: plain text concatenation (no chat template)
+- Plain format: raw instr_key + resp_key fields
+
+v2 changelog:
+  - Same token set: unconditional pass prefixes resp with a plain-text sentinel
+    (default ``\\n``; use ``--sentinel_text ""`` for bos/pad fallback).
+    Both branches predict the identical N resp tokens.
+    Single-token answers (rl=1) are now supported.
+  - ctx_len tracked in output
+  - skip_reason for None samples (no more silent None)
+  - --per_token for per-token IFD breakdown
+"""
+
+import argparse
+import json
+import statistics
+
+import torch
+import torch.nn.functional as F
+import tqdm
+
+from astrai.model import AutoModel
+from astrai.tokenize import AutoTokenizer
+
+
+def _pack_bins(pairs, max_len):
+    """BFD bin packing: pack (c+r) into bins of max total length."""
+    indexed = sorted(enumerate(pairs), key=lambda x: -(len(x[1][0]) + len(x[1][1])))
+    bins = []
+    lengths = []
+    for orig_idx, (c, r) in indexed:
+        size = len(c) + len(r)
+        best_bin = -1
+        for bi, rem in enumerate(lengths):
+            if rem >= size:
+                if best_bin < 0 or rem < lengths[best_bin]:
+                    best_bin = bi
+        if best_bin >= 0:
+            bins[best_bin].append((orig_idx, c, r))
+            lengths[best_bin] -= size
+        else:
+            bins.append([(orig_idx, c, r)])
+            lengths.append(max_len - size)
+    return bins
+
+
+def _resolve_sentinel_ids(tokenizer, sentinel_text):
+    """Tokenize the sentinel text for the unconditional pass prefix.
+
+    Falls back to bos/pad_token_id when sentinel_text is empty or
+    cannot be encoded.
+    """
+    if sentinel_text:
+        ids = tokenizer.encode(sentinel_text, add_special_tokens=False)
+        if ids:
+            return ids
+    for attr in ("bos_token_id", "pad_token_id", "eos_token_id"):
+        tid = getattr(tokenizer, attr, None)
+        if tid is not None:
+            return [tid]
+    return [0]
+
+
+@torch.inference_mode()
+def _score_batch(
+    pairs, model, device, max_len=2048, sentinel_ids=None, per_token=False
+):
+    """BFD-packed IFD with text-sentinel-anchored unconditional pass.
+
+    Conditional:   (ctx + resp[0..i-1]) → resp[i],  i = 0..N-1
+    Unconditional: (<sentinel> + resp[0..i-1]) → resp[i],  i = 0..N-1
+
+    Both branches predict the identical N response tokens.  A short
+    plain-text sentinel gives the unconditional pass a prefix so that
+    every response token can be predicted.  Single-token answers (rl=1)
+    are supported.
+    """
+    if not pairs:
+        return []
+
+    if sentinel_ids is None:
+        sentinel_ids = [0]
+
+    bins = _pack_bins(pairs, max_len)
+    result = [None] * len(pairs)
+
+    # ---- conditional pass (packed, per-document position IDs) ----
+    for bin_items in bins:
+        seq_ids = []
+        global_pos = []
+        doc_ids = []
+        doc_offsets = []
+
+        for di, (orig_idx, c, r) in enumerate(bin_items):
+            ctx_len = len(c)
+            start = len(seq_ids)
+            item_len = len(c) + len(r)
+            seq_ids.extend(c)
+            seq_ids.extend(r)
+            end = len(seq_ids)
+            global_pos.extend(range(item_len))
+            doc_ids.extend([di] * item_len)
+            doc_offsets.append((start, end, orig_idx, ctx_len))
+
+        full_ids = torch.tensor([seq_ids], device=device, dtype=torch.long)
+        pos_ids = torch.tensor([global_pos], device=device, dtype=torch.long)
+        seq_len = len(seq_ids)
+        causal = torch.tril(
+            torch.ones(seq_len, seq_len, dtype=torch.bool, device=device)
+        )
+        doc_t = torch.tensor([doc_ids], device=device)
+        doc_mask = doc_t.unsqueeze(-1) == doc_t.unsqueeze(-2)
+        attn_mask = (causal & doc_mask[0]).unsqueeze(0).unsqueeze(0)
+        logits_full = model(full_ids, position_ids=pos_ids, input_mask=attn_mask)[
+            "logits"
+        ][0]
+
+        for start, end, orig_idx, ctx_len in doc_offsets:
+            rl = end - start - ctx_len
+            resp_start = start + ctx_len - 1
+            resp_logits = logits_full[resp_start : end - 1]
+            resp_targets = torch.tensor(
+                seq_ids[start + ctx_len : end], device=device, dtype=torch.long
+            )
+            cond_losses = F.cross_entropy(
+                resp_logits, resp_targets, reduction="none"
+            ).cpu()
+            result[orig_idx] = {
+                "_cond_losses": cond_losses,
+                "_rl": rl,
+                "_ctx_len": ctx_len,
+            }
+
+    # ---- unconditional pass (sentinel-prefixed, batched 2D) ----
+    valid_items = [
+        (
+            i,
+            result[i]["_rl"],
+            result[i]["_ctx_len"],
+            result[i]["_cond_losses"],
+            pairs[i][1],
+        )
+        for i in range(len(pairs))
+        if result[i] is not None and "_cond_losses" in result[i]
+    ]
+    if not valid_items:
+        return result
+
+    valid_items.sort(key=lambda x: -x[1])
+    prefix_len = len(sentinel_ids)
+    max_rl = prefix_len + max(rl for _, rl, _, _, _ in valid_items)
+    bsz = len(valid_items)
+
+    u_batch = torch.zeros(bsz, max_rl, dtype=torch.long, device=device)
+    for ri, (_, rl, _, _, r_ids) in enumerate(valid_items):
+        u_batch[ri, :prefix_len] = torch.tensor(sentinel_ids, dtype=torch.long)
+        u_batch[ri, prefix_len : prefix_len + rl] = torch.tensor(
+            r_ids, dtype=torch.long
+        )
+
+    logits_resp = model(u_batch)["logits"]
+
+    for ri, (orig_idx, rl, ctx_len, cond_losses, _) in enumerate(valid_items):
+        unp_logits = logits_resp[ri, prefix_len - 1 : prefix_len - 1 + rl]
+        unp_targets = u_batch[ri, prefix_len : prefix_len + rl]
+        uncond_losses = F.cross_entropy(unp_logits, unp_targets, reduction="none").cpu()
+
+        L_cond = cond_losses.mean().item()
+        L_uncond = uncond_losses.mean().item()
+        ifd = L_cond / L_uncond if L_uncond > 0 else None
+
+        out = {
+            "L_cond": round(L_cond, 6),
+            "L_uncond": round(L_uncond, 6),
+            "ifd": round(ifd, 6) if ifd is not None else None,
+            "ctx_len": ctx_len,
+            "resp_len": rl,
+        }
+        if per_token:
+            per = [
+                (round(c.item() / u.item(), 6) if u.item() > 0 else None)
+                for c, u in zip(cond_losses, uncond_losses)
+            ]
+            out["ifd_per_token"] = per
+        result[orig_idx] = out
+
+    return result
+
+
+def _trim(context_ids, resp_ids, max_len):
+    """Truncate to fit max_len, keeping response intact if possible."""
+    if len(resp_ids) > max_len // 2:
+        resp_ids = resp_ids[: max_len // 2]
+    full_ids = context_ids + resp_ids
+    if len(full_ids) <= max_len:
+        return context_ids, resp_ids
+    overflow = len(full_ids) - max_len
+    if overflow >= len(context_ids):
+        return [], resp_ids[:max_len]
+    return context_ids[overflow:], resp_ids
+
+
+def score_plain(
+    model,
+    tokenizer,
+    instruction,
+    response,
+    device,
+    max_len=2048,
+    sentinel_ids=None,
+    per_token=False,
+):
+    """Compute IFD for a single instruction-response pair (plain format)."""
+    ctx_ids = tokenizer.encode(instruction, add_special_tokens=False)
+    resp_ids = tokenizer.encode(response, add_special_tokens=False)
+    ctx_ids, resp_ids = _trim(ctx_ids, resp_ids, max_len)
+    if not ctx_ids or not resp_ids:
+        return {
+            "L_cond": None,
+            "L_uncond": None,
+            "ifd": None,
+            "skip_reason": "empty ctx or resp",
+        }
+    return _score_batch(
+        [(ctx_ids, resp_ids)],
+        model,
+        device,
+        max_len,
+        sentinel_ids=sentinel_ids,
+        per_token=per_token,
+    )[0]
+
+
+def score_messages(
+    model, tokenizer, messages, device, max_len=2048, sentinel_ids=None, per_token=False
+):
+    """Compute IFD for each assistant turn in a messages array."""
+    turns = []
+    for i, msg in enumerate(messages):
+        if msg.get("role") != "assistant":
+            continue
+        ctx_text = "\n\n".join(m["content"] for m in messages[:i])
+        ctx_ids = tokenizer.encode(ctx_text)
+        resp_ids = tokenizer.encode(msg["content"], add_special_tokens=False)
+        ctx_ids, resp_ids = _trim(ctx_ids, resp_ids, max_len)
+        if ctx_ids and resp_ids:
+            turns.append((ctx_ids, resp_ids))
+    if not turns:
+        return None
+    raw_scores = _score_batch(
+        turns, model, device, max_len, sentinel_ids=sentinel_ids, per_token=per_token
+    )
+    valid = [s for s in raw_scores if s is not None and s.get("ifd") is not None]
+    if not valid:
+        return {"ifd": None, "ifd_turns": raw_scores}
+    avg = sum(s["ifd"] for s in valid) / len(valid)
+    return {
+        "ifd": avg,
+        "ifd_detail": valid[0] if len(valid) == 1 else None,
+        "ifd_turns": raw_scores,
+    }
+
+
+def process_file(
+    param_path,
+    input_file,
+    output_file,
+    instr_key,
+    resp_key,
+    max_len=2048,
+    data_format="plain",
+    batch_size=1,
+    device=None,
+    sentinel_text="\n",
+    per_token=False,
+):
+    if device is None:
+        device = "cuda" if torch.cuda.is_available() else "cpu"
+    dtype = torch.bfloat16 if "cuda" in device else torch.float32
+
+    model = AutoModel.from_pretrained(param_path)
+    tokenizer = AutoTokenizer.from_pretrained(param_path)
+    model.to(device=device, dtype=dtype)
+    model.eval()
+
+    sentinel_ids = _resolve_sentinel_ids(tokenizer, sentinel_text)
+
+    with open(input_file, encoding="utf-8") as f:
+        data = [json.loads(line) for line in f if line.strip()]
+
+    results = []
+    all_ifds = []
+    buffer = []
+
+    for item in tqdm.tqdm(data, desc="Computing IFD", unit="sample"):
+        if data_format == "messages":
+            turns = []
+            for i, msg in enumerate(item.get("messages", [])):
+                if msg.get("role") != "assistant":
+                    continue
+                ctx_text = "\n\n".join(m["content"] for m in item["messages"][:i])
+                ctx_ids = tokenizer.encode(ctx_text)
+                resp_ids = tokenizer.encode(msg["content"], add_special_tokens=False)
+                ctx_ids, resp_ids = _trim(ctx_ids, resp_ids, max_len)
+                if ctx_ids and resp_ids:
+                    turns.append((ctx_ids, resp_ids))
+            if not turns:
+                results.append(
+                    {
+                        **item,
+                        "ifd": None,
+                        "skip_reason": "no valid assistant turns",
+                        "ifd_turns": [],
+                    }
+                )
+                continue
+            buffer.append((item, turns, "messages"))
+        else:
+            ctx_ids = tokenizer.encode(item[instr_key], add_special_tokens=False)
+            resp_ids = tokenizer.encode(item[resp_key], add_special_tokens=False)
+            ctx_ids, resp_ids = _trim(ctx_ids, resp_ids, max_len)
+            if not ctx_ids or not resp_ids:
+                results.append(
+                    {
+                        **item,
+                        "ifd": None,
+                        "ifd_detail": {"skip_reason": "empty ctx or resp"},
+                    }
+                )
+                continue
+            buffer.append((item, [(ctx_ids, resp_ids)], "plain"))
+
+        if len(buffer) >= batch_size:
+            _flush_buffer(
+                buffer,
+                results,
+                all_ifds,
+                model,
+                device,
+                max_len,
+                sentinel_ids,
+                per_token,
+            )
+
+    if buffer:
+        _flush_buffer(
+            buffer, results, all_ifds, model, device, max_len, sentinel_ids, per_token
+        )
+
+    with open(output_file, "w", encoding="utf-8") as f:
+        for item in results:
+            f.write(json.dumps(item, ensure_ascii=False) + "\n")
+
+    valid_ifd = [v for v in all_ifds if v is not None]
+    if valid_ifd:
+        print(f"\n{'=' * 50}")
+        print(f"  Samples:       {len(data)}")
+        print(f"  Valid IFD:     {len(valid_ifd)}")
+        print(f"  Skipped:       {len(data) - len(valid_ifd)}")
+        print(f"  Mean IFD:      {statistics.mean(valid_ifd):.4f}")
+        print(f"  Median IFD:    {statistics.median(valid_ifd):.4f}")
+        if len(valid_ifd) > 1:
+            print(f"  Stdev IFD:     {statistics.stdev(valid_ifd):.4f}")
+        print(f"  Min IFD:       {min(valid_ifd):.4f}")
+        print(f"  Max IFD:       {max(valid_ifd):.4f}")
+        print(f"{'=' * 50}")
+    print(f"Results saved to {output_file}")
+
+
+def _flush_buffer(
+    buffer, results, all_ifds, model, device, max_len, sentinel_ids, per_token
+):
+    all_pairs = []
+    indices = []
+    for item, turns, fmt in buffer:
+        start = len(all_pairs)
+        all_pairs.extend(turns)
+        indices.append((item, turns, fmt, start, len(all_pairs)))
+
+    raw = _score_batch(
+        all_pairs,
+        model,
+        device,
+        max_len,
+        sentinel_ids=sentinel_ids,
+        per_token=per_token,
+    )
+
+    for item, turns, fmt, start, end in indices:
+        turn_scores = raw[start:end]
+        if fmt == "messages":
+            valid = [
+                s for s in turn_scores if s is not None and s.get("ifd") is not None
+            ]
+            if not valid:
+                results.append({**item, "ifd": None, "ifd_turns": turn_scores})
+            else:
+                avg = sum(s["ifd"] for s in valid) / len(valid)
+                all_ifds.append(avg)
+                results.append(
+                    {
+                        **item,
+                        "ifd": avg,
+                        "ifd_detail": valid[0] if len(valid) == 1 else None,
+                        "ifd_turns": turn_scores,
+                    }
+                )
+        else:
+            score = turn_scores[0]
+            all_ifds.append(score.get("ifd"))
+            results.append({**item, "ifd": score.get("ifd"), "ifd_detail": score})
+
+    buffer.clear()
+
+
+def main():
+    parser = argparse.ArgumentParser(
+        description="Compute IFD scores for instruction-response data"
+    )
+    parser.add_argument("--param_path", type=str, required=True, help="Model directory")
+    parser.add_argument("--input", type=str, required=True, help="Input JSONL file")
+    parser.add_argument("--output", type=str, required=True, help="Output JSONL file")
+    parser.add_argument("--max_len", type=int, default=2048, help="Max token length")
+    parser.add_argument(
+        "--format",
+        type=str,
+        default="plain",
+        choices=["plain", "messages"],
+        help="Input format",
+    )
+    parser.add_argument(
+        "--instr_key", type=str, default="instruction", help="Key for instruction field"
+    )
+    parser.add_argument(
+        "--resp_key", type=str, default="response", help="Key for response field"
+    )
+    parser.add_argument(
+        "--batch_size", type=int, default=8, help="Batch size for model forward passes"
+    )
+    parser.add_argument("--device", type=str, default=None, help="Device (e.g. cuda:0)")
+    parser.add_argument(
+        "--sentinel_text",
+        type=str,
+        default="\n",
+        help='Plain-text prefix for unconditional pass (default: "\\n"). Use "" for bos/pad fallback.',
+    )
+    parser.add_argument(
+        "--per_token",
+        action="store_true",
+        help="Include per-token IFD breakdown in output",
+    )
+    args = parser.parse_args()
+
+    process_file(
+        args.param_path,
+        args.input,
+        args.output,
+        args.instr_key,
+        args.resp_key,
+        args.max_len,
+        data_format=args.format,
+        batch_size=args.batch_size,
+        device=args.device,
+        sentinel_text=args.sentinel_text,
+        per_token=args.per_token,
+    )
+
+
+if __name__ == "__main__":
+    main()

scripts/eval/evaluate_ifeval.py

@@ -0,0 +1,609 @@
+"""IFEval instruction-following evaluation benchmark.
+
+Evaluates model responses against regex-based constraint verifiers.
+Supports all IFEval constraint types except language detection.
+
+Usage::
+
+    python scripts/tools/evaluate_ifeval.py --param_path ./params \
+        --data_path ifeval.jsonl --output results.json \
+        --temperature 0.1 --max_tokens 512
+"""
+
+import argparse
+import json
+import os
+import re
+import urllib.request
+from typing import Callable, Dict, List, Optional
+
+import torch
+import tqdm
+
+from astrai.inference import InferenceEngine
+from astrai.model import AutoModel
+from astrai.tokenize import AutoTokenizer
+
+IFEVAL_URL = (
+    "https://raw.githubusercontent.com/google-research/"
+    "google-research/master/instruction_following_eval/data/input_data.jsonl"
+)
+
+CONSTRAINT_VERIFIERS: Dict[str, Callable[[str, dict], bool]] = {}
+
+
+def register(instruction_id: str):
+    def decorator(fn):
+        CONSTRAINT_VERIFIERS[instruction_id] = fn
+        return fn
+
+    return decorator
+
+
+@register("keywords:existence")
+def check_keyword_existence(response: str, kwargs: dict) -> bool:
+    for kw in kwargs["keywords"]:
+        if not re.search(re.escape(kw), response, re.IGNORECASE):
+            return False
+    return True
+
+
+@register("keywords:frequency")
+def check_keyword_frequency(response: str, kwargs: dict) -> bool:
+    keyword = kwargs["keyword"]
+    frequency = kwargs.get("frequency", 1)
+    relation = kwargs.get("relation", "at least")
+    count = len(re.findall(re.escape(keyword), response, re.IGNORECASE))
+    if relation == "less than":
+        return count < frequency
+    return count >= frequency
+
+
+@register("keywords:forbidden_words")
+def check_forbidden_words(response: str, kwargs: dict) -> bool:
+    for word in kwargs["forbidden_words"]:
+        if re.search(r"\b" + re.escape(word) + r"\b", response, re.IGNORECASE):
+            return False
+    return True
+
+
+@register("keywords:letter_frequency")
+def check_letter_frequency(response: str, kwargs: dict) -> bool:
+    letter = kwargs["letter"].lower()
+    frequency = kwargs.get("let_frequency", 1)
+    relation = kwargs.get("let_relation", "at least")
+    count = response.lower().count(letter)
+    if relation == "less than":
+        return count < frequency
+    return count >= frequency
+
+
+@register("detectable_content:number_placeholders")
+def check_placeholders(response: str, kwargs: dict) -> bool:
+    num = kwargs.get("num_placeholders", 1)
+    placeholders = re.findall(r"\[.*?\]", response)
+    return len(placeholders) >= num
+
+
+@register("detectable_content:postscript")
+def check_postscript(response: str, kwargs: dict) -> bool:
+    marker = kwargs.get("postscript_marker", "P.S.")
+    response_lower = response.lower()
+    if marker == "P.P.S":
+        return bool(re.search(r"p\.\s?p\.\s?s", response_lower))
+    elif marker == "P.S.":
+        return bool(re.search(r"p\.\s?s\.", response_lower))
+    else:
+        return bool(re.search(re.escape(marker.lower()), response_lower))
+
+
+@register("detectable_format:number_bullet_lists")
+def check_bullet_lists(response: str, kwargs: dict) -> bool:
+    num = kwargs.get("num_bullets", 1)
+    bullets = re.findall(r"^\s*\*[^\*].*$", response, re.MULTILINE)
+    dashes = re.findall(r"^\s*-.*$", response, re.MULTILINE)
+    return len(bullets) + len(dashes) == num
+
+
+@register("detectable_format:number_highlighted_sections")
+def check_highlighted_sections(response: str, kwargs: dict) -> bool:
+    num = kwargs.get("num_highlights", 1)
+    highlights = re.findall(r"\*[^\n\*]+\*", response)
+    count = 0
+    for h in highlights:
+        if h.strip("*").strip():
+            count += 1
+    return count >= num
+
+
+@register("detectable_format:multiple_sections")
+def check_multiple_sections(response: str, kwargs: dict) -> bool:
+    splitter = kwargs.get("section_spliter", "Section")
+    num = kwargs.get("num_sections", 1)
+    pattern = r"\s?" + re.escape(splitter) + r"\s?\d+\s?"
+    sections = re.split(pattern, response)
+    return len(sections) - 1 >= num
+
+
+@register("detectable_format:title")
+def check_title(response: str, kwargs: dict) -> bool:
+    titles = re.findall(r"<<[^>\n]+>>", response)
+    for title in titles:
+        if title.strip("<>").strip():
+            return True
+    return False
+
+
+@register("detectable_format:json_format")
+def check_json_format(response: str, kwargs: dict) -> bool:
+    value = response.strip()
+    for prefix in ("```json", "```Json", "```JSON", "```"):
+        if value.lower().startswith(prefix.lower()):
+            value = value[len(prefix) :].strip()
+    if value.endswith("```"):
+        value = value[:-3].strip()
+    try:
+        json.loads(value)
+        return True
+    except (ValueError, json.JSONDecodeError):
+        return False
+
+
+@register("detectable_format:general_punctuation")
+def check_general_punctuation(response: str, kwargs: dict) -> bool:
+    punctuation_blacklist = kwargs.get("punctuation_blacklist", [])
+    for punct in punctuation_blacklist:
+        if punct in response:
+            return False
+    return True
+
+
+@register("detectable_format:number_highlighted_words")
+def check_highlighted_words(response: str, kwargs: dict) -> bool:
+    num = kwargs.get("num_highlights", 1)
+    highlights = re.findall(r"\*[^\s\*][^\*]*[^\s\*]\*", response)
+    return len(highlights) >= num
+
+
+@register("startend:end_checker")
+def check_end_checker(response: str, kwargs: dict) -> bool:
+    end_phrase = kwargs["end_phrase"]
+    return (
+        response.strip()
+        .rstrip('"')
+        .rstrip()
+        .lower()
+        .endswith(end_phrase.strip().lower())
+    )
+
+
+@register("startend:quotation")
+def check_quotation(response: str, kwargs: dict) -> bool:
+    value = response.strip()
+    return value.startswith('"') and value.endswith('"')
+
+
+@register("startend:start_checker")
+def check_start_checker(response: str, kwargs: dict) -> bool:
+    starter = kwargs["starter"]
+    return bool(re.search(r"^\s*" + re.escape(starter), response, re.MULTILINE))
+
+
+@register("change_case:english_capital")
+def check_english_capital(response: str, kwargs: dict) -> bool:
+    return response.isupper()
+
+
+@register("change_case:english_lowercase")
+def check_english_lowercase(response: str, kwargs: dict) -> bool:
+    return response.islower()
+
+
+@register("change_case:capital_word_frequency")
+def check_capital_word_frequency(response: str, kwargs: dict) -> bool:
+    frequency = kwargs.get("capital_frequency", 1)
+    relation = kwargs.get("capital_relation", "at least")
+    capital_words = re.findall(r"\b[A-Z]{2,}\b", response)
+    count = len(capital_words)
+    if relation == "less than":
+        return count < frequency
+    return count >= frequency
+
+
+@register("punctuation:no_comma")
+def check_no_comma(response: str, kwargs: dict) -> bool:
+    return "," not in response
+
+
+def count_words(text: str) -> int:
+    return len(re.findall(r"\b\w+\b", text))
+
+
+def count_sentences(text: str) -> int:
+    text = text.strip()
+    if not text:
+        return 0
+    sentences = re.split(r"(?<=[.!?])\s+", text)
+    return len([s for s in sentences if s.strip()])
+
+
+@register("length_constraints:number_words")
+def check_number_words(response: str, kwargs: dict) -> bool:
+    num = kwargs.get("num_words", 100)
+    relation = kwargs.get("relation", "at least")
+    cnt = count_words(response)
+    if relation == "less than":
+        return cnt < num
+    return cnt >= num
+
+
+@register("length_constraints:number_sentences")
+def check_number_sentences(response: str, kwargs: dict) -> bool:
+    num = kwargs.get("num_sentences", 5)
+    relation = kwargs.get("relation", "at least")
+    cnt = count_sentences(response)
+    if relation == "less than":
+        return cnt < num
+    return cnt >= num
+
+
+@register("length_constraints:number_paragraphs")
+def check_number_paragraphs(response: str, kwargs: dict) -> bool:
+    num = kwargs.get("num_paragraphs", 1)
+    if "***" in response:
+        paragraphs = re.split(r"\s?\*\*\*\s?", response)
+    else:
+        paragraphs = re.split(r"\n\n+", response)
+    actual = len([p for p in paragraphs if p.strip()])
+    return actual == num
+
+
+@register("length_constraints:nth_paragraph_first_word")
+def check_nth_paragraph_first_word(response: str, kwargs: dict) -> bool:
+    num_paragraphs = kwargs.get("num_paragraphs", 1)
+    nth = kwargs.get("nth_paragraph", 1)
+    first_word = kwargs.get("first_word", "").lower()
+
+    paragraphs = re.split(r"\n\n+", response)
+    paragraphs = [p.strip() for p in paragraphs if p.strip()]
+
+    if len(paragraphs) != num_paragraphs:
+        return False
+    if nth > len(paragraphs):
+        return False
+
+    target = paragraphs[nth - 1]
+    words = target.split()
+    if not words:
+        return False
+
+    word = words[0].strip().lstrip("'\"").rstrip(".,!?:;\"'")
+    return word.lower() == first_word
+
+
+@register("length_constraints:nth_word_checker")
+def check_nth_word(response: str, kwargs: dict) -> bool:
+    nth = kwargs.get("nth_word", 1)
+    target = kwargs.get("target_word", "").lower()
+    words = re.findall(r"\b\w+\b", response)
+    if nth > len(words):
+        return False
+    return words[nth - 1].lower() == target
+
+
+@register("combination:repeat_prompt")
+def check_repeat_prompt(response: str, kwargs: dict) -> bool:
+    prompt = kwargs["prompt_to_repeat"]
+    return response.strip().lower().startswith(prompt.strip().lower())
+
+
+@register("combination:two_responses")
+def check_two_responses(response: str, kwargs: dict) -> bool:
+    parts = response.split("******")
+    valid = [p for p in parts if p.strip()]
+    if len(valid) != 2:
+        return False
+    return valid[0].strip() != valid[1].strip()
+
+
+def download_ifeval(data_path: str):
+    if os.path.exists(data_path):
+        return
+    os.makedirs(os.path.dirname(data_path) or ".", exist_ok=True)
+    print(f"Downloading IFEval from {IFEVAL_URL} ...")
+    tmp = data_path + ".tmp"
+    urllib.request.urlretrieve(IFEVAL_URL, tmp)
+    with open(tmp, "rb") as f_in:
+        content = f_in.read()
+    with open(data_path, "wb") as f_out:
+        f_out.write(content)
+    os.remove(tmp)
+    print(f"  saved to {data_path}")
+
+
+def load_problems(data_path: str) -> List[dict]:
+    problems = []
+    with open(data_path, "r", encoding="utf-8") as f:
+        for line in f:
+            line = line.strip()
+            if line:
+                problems.append(json.loads(line))
+    return problems
+
+
+def verify_response(response: str, instruction_id: str, kwargs: dict) -> Optional[bool]:
+    verifier = CONSTRAINT_VERIFIERS.get(instruction_id)
+    if verifier is None:
+        return None
+    try:
+        return verifier(response, kwargs)
+    except Exception:
+        return False
+
+
+def generate_one(
+    engine: InferenceEngine,
+    tokenizer: AutoTokenizer,
+    prompt: str,
+    max_tokens: int,
+    temperature: float,
+    top_p: float,
+    top_k: int,
+) -> str:
+    formatted = tokenizer.apply_chat_template(
+        [{"role": "user", "content": prompt}],
+        tokenize=False,
+        add_generation_prompt=True,
+    )
+    output = engine.generate(
+        prompt=formatted,
+        stream=False,
+        max_tokens=max_tokens,
+        temperature=temperature,
+        top_p=top_p,
+        top_k=top_k,
+    )
+    if isinstance(output, list):
+        return output[0]
+    return output
+
+
+def evaluate(
+    engine: InferenceEngine,
+    tokenizer: AutoTokenizer,
+    problems: List[dict],
+    max_tokens: int,
+    temperature: float,
+    top_p: float,
+    top_k: int,
+    num_samples: int = 1,
+) -> Dict:
+    results = {}
+    constraint_stats: Dict[str, Dict[str, int]] = {}
+    total_constraints = 0
+    total_passed = 0
+
+    for problem in tqdm.tqdm(problems, desc="IFEval", unit="problem"):
+        key = problem["key"]
+        prompt = problem["prompt"]
+        instruction_ids = problem["instruction_id_list"]
+        kwargs_list = problem["kwargs"]
+
+        samples = []
+        for _ in range(num_samples):
+            response = generate_one(
+                engine, tokenizer, prompt, max_tokens, temperature, top_p, top_k
+            )
+            samples.append(response)
+
+        constraint_results = []
+        passed = 0
+        verified = 0
+
+        for idx, instruction_id in enumerate(instruction_ids):
+            kwargs = kwargs_list[idx] if idx < len(kwargs_list) else {}
+            best_pass = False
+            for response in samples:
+                result = verify_response(response, instruction_id, kwargs)
+                if result is None:
+                    continue
+                if result:
+                    best_pass = True
+                    break
+
+            verifier_exists = instruction_id in CONSTRAINT_VERIFIERS
+            if verifier_exists:
+                verified += 1
+                if best_pass:
+                    passed += 1
+
+            constraint_results.append(
+                {
+                    "instruction_id": instruction_id,
+                    "passed": best_pass,
+                    "supported": verifier_exists,
+                    "kwargs": kwargs,
+                }
+            )
+
+            if verifier_exists:
+                if instruction_id not in constraint_stats:
+                    constraint_stats[instruction_id] = {
+                        "total": 0,
+                        "passed": 0,
+                    }
+                constraint_stats[instruction_id]["total"] += 1
+                if best_pass:
+                    constraint_stats[instruction_id]["passed"] += 1
+
+        total_constraints += verified
+        total_passed += passed
+
+        accuracy = passed / verified if verified > 0 else None
+        results[str(key)] = {
+            "key": key,
+            "prompt": prompt,
+            "response": samples[0],
+            "num_samples": num_samples,
+            "num_constraints": len(instruction_ids),
+            "num_verified": verified,
+            "num_passed": passed,
+            "accuracy": round(accuracy, 4) if accuracy is not None else None,
+            "constraints": constraint_results,
+        }
+
+    overall_accuracy = (
+        round(total_passed / total_constraints, 4) if total_constraints > 0 else 0.0
+    )
+
+    type_summary = {}
+    for inst_id, stats in sorted(constraint_stats.items()):
+        type_summary[inst_id] = {
+            "total": stats["total"],
+            "passed": stats["passed"],
+            "accuracy": round(stats["passed"] / stats["total"], 4)
+            if stats["total"] > 0
+            else 0.0,
+        }
+
+    unsupported_count = sum(
+        1
+        for p in problems
+        for iid in p["instruction_id_list"]
+        if iid not in CONSTRAINT_VERIFIERS
+    )
+
+    results["_summary"] = {
+        "total_problems": len(problems),
+        "total_constraints": total_constraints,
+        "total_passed": total_passed,
+        "overall_accuracy": overall_accuracy,
+        "unsupported_constraints": unsupported_count,
+        "supported_types": sorted(CONSTRAINT_VERIFIERS.keys()),
+        "per_type_accuracy": type_summary,
+    }
+
+    return results
+
+
+def main():
+    parser = argparse.ArgumentParser(description="IFEval benchmark")
+    parser.add_argument(
+        "--param_path", type=str, default="./params", help="Model directory"
+    )
+    parser.add_argument(
+        "--data_path",
+        type=str,
+        default="./ifeval/input_data.jsonl",
+        help="IFEval JSONL file (auto-download if missing)",
+    )
+    parser.add_argument("--output", type=str, default=None, help="Output JSON path")
+    parser.add_argument(
+        "--max_tokens", type=int, default=512, help="Max generation tokens"
+    )
+    parser.add_argument(
+        "--temperature",
+        type=float,
+        default=0.1,
+        help="Sampling temperature",
+    )
+    parser.add_argument("--top_p", type=float, default=0.95, help="Top-p sampling")
+    parser.add_argument("--top_k", type=int, default=50, help="Top-k sampling")
+    parser.add_argument(
+        "--num_samples",
+        type=int,
+        default=1,
+        help="Number of samples per problem (best-of-n scoring)",
+    )
+    parser.add_argument(
+        "--batch_size", type=int, default=1, help="Inference batch size"
+    )
+    parser.add_argument(
+        "--limit",
+        type=int,
+        default=None,
+        help="Limit to first N problems (for quick testing)",
+    )
+    parser.add_argument(
+        "--dump_responses",
+        type=str,
+        default=None,
+        help="Path to dump raw model responses (JSONL)",
+    )
+    args = parser.parse_args()
+
+    download_ifeval(args.data_path)
+    problems = load_problems(args.data_path)
+    if args.limit:
+        problems = problems[: args.limit]
+
+    print(f"Loaded {len(problems)} problems")
+    print(f"Supported constraint types: {len(CONSTRAINT_VERIFIERS)}")
+
+    model = AutoModel.from_pretrained(args.param_path)
+    tokenizer = AutoTokenizer.from_pretrained(args.param_path)
+    model.to(device="cuda", dtype=torch.bfloat16)
+    model.eval()
+
+    engine = InferenceEngine(
+        model=model,
+        tokenizer=tokenizer,
+        max_batch_size=args.batch_size,
+    )
+
+    results = evaluate(
+        engine=engine,
+        tokenizer=tokenizer,
+        problems=problems,
+        max_tokens=args.max_tokens,
+        temperature=args.temperature,
+        top_p=args.top_p,
+        top_k=args.top_k,
+        num_samples=args.num_samples,
+    )
+
+    summary = results.pop("_summary")
+    print(f"\n{'=' * 60}")
+    print(f"  Problems:       {summary['total_problems']}")
+    print(f"  Constraints:    {summary['total_constraints']}")
+    print(f"  Passed:         {summary['total_passed']}")
+    print(f"  Accuracy:       {summary['overall_accuracy']:.2%}")
+    print(f"  Unsupported:    {summary['unsupported_constraints']}")
+    print(f"{'=' * 60}")
+
+    print("\nPer-type accuracy:")
+    for inst_id, stats in sorted(summary["per_type_accuracy"].items()):
+        print(
+            f"  {inst_id:50s}  {stats['accuracy']:.2%}  "
+            f"({stats['passed']}/{stats['total']})"
+        )
+
+    if args.output:
+        results["_summary"] = summary
+        with open(args.output, "w", encoding="utf-8") as f:
+            json.dump(results, f, indent=2, ensure_ascii=False)
+        print(f"\nResults saved to {args.output}")
+
+    if args.dump_responses:
+        with open(args.dump_responses, "w", encoding="utf-8") as f:
+            for k, v in results.items():
+                if k.startswith("_"):
+                    continue
+                f.write(
+                    json.dumps(
+                        {
+                            "key": v["key"],
+                            "prompt": v["prompt"],
+                            "response": v["response"],
+                        },
+                        ensure_ascii=False,
+                    )
+                    + "\n"
+                )
+        print(f"Responses dumped to {args.dump_responses}")
+
+    engine.shutdown()
+
+
+if __name__ == "__main__":
+    main()

scripts/eval/evaluate_mmlu.py

@@ -5,9 +5,9 @@ import csv
 import json
 import os
 import shutil
-import urllib.request
-import zipfile
+import tarfile
 
+import requests
 import torch
 import torch.nn.functional as F
 import tqdm
@@ -15,7 +15,7 @@ import tqdm
 from astrai.model import AutoModel
 from astrai.tokenize import AutoTokenizer
 
-MMLU_URL = "https://github.com/hendrycks/test/archive/refs/heads/master.zip"
+MMLU_URL = "https://people.eecs.berkeley.edu/~hendrycks/data.tar"
 MMLU_SUBJECTS = [
     "abstract_algebra",
     "anatomy",
@@ -78,23 +78,37 @@ MMLU_SUBJECTS = [
 
 
 def _download_and_extract(url: str, data_dir: str):
-    zip_path = os.path.join(data_dir, "mmlu.zip")
+    tar_path = os.path.join(data_dir, "data.tar")
     os.makedirs(data_dir, exist_ok=True)
     print(f"Downloading MMLU data from {url}...")
-    urllib.request.urlretrieve(url, zip_path)
+    resp = requests.get(url, stream=True, timeout=300)
+    resp.raise_for_status()
+    total = int(resp.headers.get("content-length", 0))
+    with tqdm.tqdm(total=total, unit="B", unit_scale=True, desc="  Download") as bar:
+        with open(tar_path, "wb") as f:
+            for chunk in resp.iter_content(chunk_size=8192):
+                f.write(chunk)
+                bar.update(len(chunk))
     print("Extracting...")
-    with zipfile.ZipFile(zip_path, "r") as zf:
-        zf.extractall(data_dir)
-    os.remove(zip_path)
+    with tarfile.open(tar_path, "r") as tf:
+        tf.extractall(data_dir)
+    os.remove(tar_path)
 
 
 def download_mmlu(data_dir: str):
     _download_and_extract(MMLU_URL, data_dir)
-    src = os.path.join(data_dir, "test-master", "data")
+    src = os.path.join(data_dir, "data")
     if os.path.exists(src):
         for item in os.listdir(src):
-            os.rename(os.path.join(src, item), os.path.join(data_dir, item))
-        shutil.rmtree(os.path.join(data_dir, "test-master"))
+            src_item = os.path.join(src, item)
+            dst_item = os.path.join(data_dir, item)
+            if os.path.exists(dst_item):
+                if os.path.isdir(dst_item):
+                    shutil.rmtree(dst_item)
+                else:
+                    os.remove(dst_item)
+            os.rename(src_item, dst_item)
+        os.rmdir(src)
     print(f"MMLU data saved to {data_dir}")
 
 
@@ -143,10 +157,32 @@ def build_prompt(
     return prompt
 
 
+def apply_chat(
+    tokenizer, raw_prompt: str, n_shot: int, dev_data: list[dict] | None
+) -> str:
+    """Wrap raw MMLU prompt in the model's chat template format.
+
+    For few-shot, prepend example Q&A pairs as a second user/assistant exchange.
+    """
+    messages = []
+    if n_shot > 0 and dev_data:
+        for item in dev_data[:n_shot]:
+            q = f"Question: {item['question']}\n"
+            for k in ("A", "B", "C", "D"):
+                q += f"{k}. {item[k]}\n"
+            q += "Answer:"
+            messages.append({"role": "user", "content": q})
+            messages.append({"role": "assistant", "content": item["answer"]})
+    messages.append({"role": "user", "content": raw_prompt})
+    return tokenizer.apply_chat_template(
+        messages, tokenize=False, add_generation_prompt=True
+    )
+
+
 def choice_logprob(
     model, tokenizer, context_ids: list[int], choice_letter: str, device: str
 ) -> float:
-    choice_text = f" {choice_letter}"
+    choice_text = choice_letter
     choice_ids = tokenizer.encode(choice_text, add_special_tokens=False)
     input_ids = context_ids + choice_ids
     max_len = model.config.max_len
@@ -182,8 +218,11 @@ def evaluate_subject(
     correct = 0
     total = 0
     for item in tqdm.tqdm(test_data, desc=f"{subject:40s}", leave=False):
-        prompt = build_prompt(item["question"], item, subject, n_shot, dev_data or [])
-        context_ids = tokenizer.encode(prompt)
+        raw_prompt = build_prompt(
+            item["question"], item, subject, n_shot, dev_data or []
+        )
+        context = apply_chat(tokenizer, raw_prompt, n_shot, dev_data or [])
+        context_ids = tokenizer.encode(context)
         scores = {
             c: choice_logprob(model, tokenizer, context_ids, c, device)
             for c in ("A", "B", "C", "D")
@@ -233,6 +272,7 @@ def main():
     device = args.device
     dtype = getattr(torch, args.dtype)
     model.to(device=device, dtype=dtype)
+    model.eval()
 
     subjects = args.subjects or MMLU_SUBJECTS
     results = {}

scripts/eval/evaluate_ppl.py

@@ -86,7 +86,7 @@ def process_file(
 
 
 if __name__ == "__main__":
-    parser = argparse.ArgumentParser(description="Run perplexity with a Khaosz model.")
+    parser = argparse.ArgumentParser(description="Perplexity evaluation on JSONL text.")
     parser.add_argument(
         "--param_path", type=str, required=True, help="Path to the model directory."
     )

scripts/eval/evaluate_rouge.py

@@ -0,0 +1,153 @@
+"""ROUGE evaluation (manual implementation, no external deps).
+
+Computes ROUGE-1, ROUGE-2, ROUGE-L precision, recall, and F1.
+
+Usage::
+
+    # Batch evaluation from JSONL (each line: {"reference": ..., "candidate": ...})
+    python scripts/eval/evaluate_rouge.py --data_path preds.jsonl --output results.json
+
+    # As a library
+    from scripts.eval.evaluate_rouge import compute_rouge
+    scores = compute_rouge("the cat sat on the mat", "the cat sat")
+"""
+
+import argparse
+import json
+from collections import Counter
+from typing import Dict, List, Tuple
+
+
+def _tokenize(text: str) -> List[str]:
+    return text.split()
+
+
+def _ngrams(tokens: List[str], n: int) -> Counter:
+    return Counter(zip(*[tokens[i:] for i in range(n)]))
+
+
+def _lcs(x: List[str], y: List[str]) -> int:
+    m, n = len(x), len(y)
+    dp = [[0] * (n + 1) for _ in range(m + 1)]
+    for i in range(1, m + 1):
+        xi = x[i - 1]
+        dpi = dp[i]
+        dpi_1 = dp[i - 1]
+        for j in range(1, n + 1):
+            if xi == y[j - 1]:
+                dpi[j] = dpi_1[j - 1] + 1
+            else:
+                dpi[j] = dpi_1[j] if dpi_1[j] > dpi[j - 1] else dpi[j - 1]
+    return dp[m][n]
+
+
+def _f1(precision: float, recall: float) -> float:
+    if precision + recall == 0:
+        return 0.0
+    return 2 * precision * recall / (precision + recall)
+
+
+def _rouge_n(ref_tokens: List[str], cand_tokens: List[str], n: int) -> Dict[str, float]:
+    ref_ngrams = _ngrams(ref_tokens, n)
+    cand_ngrams = _ngrams(cand_tokens, n)
+
+    overlap = sum((cand_ngrams & ref_ngrams).values())
+    cand_total = sum(cand_ngrams.values())
+    ref_total = sum(ref_ngrams.values())
+
+    precision = overlap / cand_total if cand_total > 0 else 0.0
+    recall = overlap / ref_total if ref_total > 0 else 0.0
+    f1 = _f1(precision, recall)
+
+    return {"precision": precision, "recall": recall, "f1": f1}
+
+
+def _rouge_l(ref_tokens: List[str], cand_tokens: List[str]) -> Dict[str, float]:
+    lcs_len = _lcs(ref_tokens, cand_tokens)
+    ref_len = len(ref_tokens)
+    cand_len = len(cand_tokens)
+
+    recall = lcs_len / ref_len if ref_len > 0 else 0.0
+    precision = lcs_len / cand_len if cand_len > 0 else 0.0
+    f1 = _f1(precision, recall)
+
+    return {"precision": precision, "recall": recall, "f1": f1}
+
+
+def compute_rouge(
+    reference: str, candidate: str, n: int = 2
+) -> Dict[str, Dict[str, float]]:
+    """Compute ROUGE-N (1..n) and ROUGE-L scores.
+
+    Returns::
+
+        {
+            "rouge-1": {"precision": ..., "recall": ..., "f1": ...},
+            "rouge-2": {"precision": ..., "recall": ..., "f1": ...},
+            "rouge-l": {"precision": ..., "recall": ..., "f1": ...},
+        }
+    """
+    ref_tokens = _tokenize(reference)
+    cand_tokens = _tokenize(candidate)
+
+    results = {}
+    for i in range(1, n + 1):
+        results[f"rouge-{i}"] = _rouge_n(ref_tokens, cand_tokens, i)
+    results["rouge-l"] = _rouge_l(ref_tokens, cand_tokens)
+    return results
+
+
+def evaluate_file(data_path: str) -> Dict:
+    with open(data_path, "r", encoding="utf-8") as f:
+        pairs = [json.loads(line) for line in f if line.strip()]
+
+    agg = {
+        k: {"precision": 0.0, "recall": 0.0, "f1": 0.0}
+        for k in ("rouge-1", "rouge-2", "rouge-l")
+    }
+    per_item = []
+
+    for item in pairs:
+        ref = item["reference"]
+        cand = item["candidate"]
+        scores = compute_rouge(ref, cand)
+        per_item.append({**item, "scores": scores})
+        for k, v in scores.items():
+            agg[k]["precision"] += v["precision"]
+            agg[k]["recall"] += v["recall"]
+            agg[k]["f1"] += v["f1"]
+
+    n = len(pairs)
+    for k in agg:
+        agg[k] = {m: v / n for m, v in agg[k].items()}
+
+    return {"num_samples": n, "aggregate": agg, "per_item": per_item}
+
+
+def main():
+    parser = argparse.ArgumentParser(description="ROUGE evaluation")
+    parser.add_argument(
+        "--data_path", required=True, help="JSONL with reference/candidate per line"
+    )
+    parser.add_argument("--output", type=str, default=None, help="Output JSON path")
+    args = parser.parse_args()
+
+    results = evaluate_file(args.data_path)
+    agg = results["aggregate"]
+
+    print(f"Samples: {results['num_samples']}")
+    print()
+    for metric in ("rouge-1", "rouge-2", "rouge-l"):
+        s = agg[metric]
+        print(
+            f"  {metric:8s}  P={s['precision']:.4f}  R={s['recall']:.4f}  F1={s['f1']:.4f}"
+        )
+
+    if args.output:
+        with open(args.output, "w", encoding="utf-8") as f:
+            json.dump(results, f, indent=2, ensure_ascii=False)
+        print(f"\nSaved to {args.output}")
+
+
+if __name__ == "__main__":
+    main()

scripts/tools/benchmark.py

@@ -1,12 +1,13 @@
 """Benchmark AutoRegressiveLM with KVCache"""
 
+import argparse
 from dataclasses import dataclass
 from typing import Any, Dict
 
 import torch
 
 from astrai.config import AutoRegressiveLMConfig
-from astrai.inference import KVCache
+from astrai.inference import ContiguousCache, PageCache
 from astrai.model.transformer import AutoRegressiveLM
 
 
@@ -24,41 +25,14 @@ class GenerationBenchmark:
         config: AutoRegressiveLMConfig,
         device: str = "cuda",
         dtype: torch.dtype = torch.bfloat16,
-        page_size: int = 128,
+        cache_type: str = "contiguous",
     ):
         self.config = config
         self.device = device
         self.dtype = dtype
+        self.cache_type = cache_type
         self.model = AutoRegressiveLM(config).to(device=device, dtype=dtype)
         self.model.eval()
-        head_dim = config.dim // config.n_heads
-        n_pages = (config.max_len * 4 + page_size - 1) // page_size
-        self._page_cache = KVCache(
-            config.n_layers,
-            n_pages,
-            page_size,
-            config.n_kv_heads,
-            head_dim,
-            device,
-            dtype,
-        )
-
-    def _prepare_inputs(self, batch_size: int, prompt_length: int, total_length: int):
-        prompt_ids = torch.randint(
-            low=0,
-            high=self.config.vocab_size,
-            size=(batch_size, prompt_length),
-            device=self.device,
-            dtype=torch.long,
-        )
-        gen_ids = torch.randint(
-            low=0,
-            high=self.config.vocab_size,
-            size=(batch_size, total_length - prompt_length),
-            device=self.device,
-            dtype=torch.long,
-        )
-        return prompt_ids, gen_ids
 
     @torch.inference_mode()
     def run_prefill_benchmark(
@@ -68,8 +42,12 @@ class GenerationBenchmark:
         num_trials: int = 10,
     ) -> BenchmarkResult:
         for _ in range(3):
-            prompt_ids, _ = self._prepare_inputs(
-                batch_size, prompt_length, prompt_length
+            prompt_ids = torch.randint(
+                0,
+                self.config.vocab_size,
+                (batch_size, prompt_length),
+                device=self.device,
+                dtype=torch.long,
             )
             _ = self.model(prompt_ids)
         torch.cuda.synchronize()
@@ -78,12 +56,15 @@ class GenerationBenchmark:
         total_tokens = batch_size * prompt_length * num_trials
 
         for trial in range(num_trials):
-            prompt_ids, _ = self._prepare_inputs(
-                batch_size, prompt_length, prompt_length
+            prompt_ids = torch.randint(
+                0,
+                self.config.vocab_size,
+                (batch_size, prompt_length),
+                device=self.device,
+                dtype=torch.long,
             )
             start = torch.cuda.Event(enable_timing=True)
             end = torch.cuda.Event(enable_timing=True)
-
             start.record()
             _ = self.model(prompt_ids)
             end.record()
@@ -107,6 +88,7 @@ class GenerationBenchmark:
                 "prompt_length": prompt_length,
                 "dtype": str(self.dtype),
                 "device": self.device,
+                "cache": "none",
             },
         )
 
@@ -120,29 +102,56 @@ class GenerationBenchmark:
     ) -> BenchmarkResult:
         total_time = 0.0
         total_tokens = batch_size * gen_length * num_trials
-        page_size = self._page_cache.page_size
 
         for trial in range(num_trials):
-            prompt_ids, gen_ids = self._prepare_inputs(
-                batch_size,
-                prompt_length,
-                prompt_length + gen_length,
-            )
-
-            n_pages = (prompt_length + gen_length + page_size - 1) // page_size
-            total = n_pages * batch_size
-            pages = []
-            for _ in range(total):
-                p = self._page_cache._pool.alloc()
-                assert p >= 0, "OOM"
-                pages.append(p)
-            page_table = torch.tensor(
-                [pages[i * n_pages : (i + 1) * n_pages] for i in range(batch_size)],
-                dtype=torch.long,
+            prompt_ids = torch.randint(
+                0,
+                self.config.vocab_size,
+                (batch_size, prompt_length),
                 device=self.device,
+                dtype=torch.long,
+            )
+            gen_ids = torch.randint(
+                0,
+                self.config.vocab_size,
+                (batch_size, gen_length),
+                device=self.device,
+                dtype=torch.long,
             )
 
-            cv = self._page_cache.bind(page_table, total_len=prompt_length)
+            head_dim = self.config.dim // self.config.n_heads
+            max_seq = prompt_length + gen_length
+
+            if self.cache_type == "contiguous":
+                cache = ContiguousCache(
+                    self.config.n_layers,
+                    batch_size,
+                    max_seq,
+                    self.config.n_kv_heads,
+                    head_dim,
+                    self.device,
+                    self.dtype,
+                )
+            else:
+                page_size = 128
+                n_pages = (max_seq + page_size - 1) // page_size * batch_size
+                cache = PageCache(
+                    self.config.n_layers,
+                    n_pages,
+                    page_size,
+                    self.config.n_kv_heads,
+                    head_dim,
+                    self.device,
+                    self.dtype,
+                )
+
+            task_ids = [f"b{i}" for i in range(batch_size)]
+            for tid in task_ids:
+                cache.task_alloc(tid, [0] * max_seq)
+                for p in range(max_seq):
+                    cache.task_extend(tid, p)
+
+            cv = cache.bind_tasks(task_ids, prompt_length, self.device)
             _ = self.model(
                 prompt_ids,
                 paged_cache=cv,
@@ -152,37 +161,35 @@ class GenerationBenchmark:
                 .unsqueeze(0)
                 .expand(batch_size, -1),
             )
-
             torch.cuda.synchronize()
 
             start = torch.cuda.Event(enable_timing=True)
             end = torch.cuda.Event(enable_timing=True)
-
             start.record()
-            current_pos = prompt_length
+
             for i in range(gen_length):
-                input_token = gen_ids[:, i : i + 1]
-                cv = self._page_cache.bind(page_table, total_len=current_pos + 1)
+                pos = prompt_length + i
+                cv = cache.bind_tasks(task_ids, pos + 1, self.device)
                 _ = self.model(
-                    input_token,
+                    gen_ids[:, i : i + 1],
                     paged_cache=cv,
                     position_ids=torch.full(
                         (batch_size, 1),
-                        current_pos,
+                        pos,
                         dtype=torch.long,
                         device=self.device,
                     ),
                 )
-                current_pos += 1
+
             end.record()
             torch.cuda.synchronize()
 
+            for tid in task_ids:
+                cache.task_free(tid)
+
             trial_time = start.elapsed_time(end) / 1000
             total_time += trial_time
 
-            for idx in pages:
-                self._page_cache._pool.free(idx)
-
             print(
                 f"  Trial {trial + 1}/{num_trials}: {gen_length} tokens in {trial_time:.3f}s "
                 f"({gen_length / trial_time:.1f} tok/s)"
@@ -199,6 +206,7 @@ class GenerationBenchmark:
                 "gen_length": gen_length,
                 "dtype": str(self.dtype),
                 "device": self.device,
+                "cache": self.cache_type,
             },
         )
 
@@ -216,6 +224,42 @@ def print_benchmark_result(result: BenchmarkResult):
 
 
 if __name__ == "__main__":
+    parser = argparse.ArgumentParser(description="AutoRegressiveLM benchmark")
+    parser.add_argument(
+        "--device", type=str, default="cuda", help="Device (default: cuda)"
+    )
+    parser.add_argument(
+        "--dtype",
+        type=str,
+        default="bfloat16",
+        choices=["bfloat16", "float16", "float32"],
+        help="Dtype",
+    )
+    parser.add_argument(
+        "--cache",
+        type=str,
+        default="contiguous",
+        choices=["contiguous", "paged"],
+        help="KV cache type",
+    )
+    parser.add_argument("--batch_size", type=int, default=4, help="Batch size")
+    parser.add_argument("--prompt_length", type=int, default=512, help="Prompt length")
+    parser.add_argument("--gen_length", type=int, default=128, help="Generation length")
+    parser.add_argument("--num_trials", type=int, default=5, help="Number of trials")
+    parser.add_argument(
+        "--prefill_only", action="store_true", help="Run prefill benchmark only"
+    )
+    parser.add_argument(
+        "--decode_only", action="store_true", help="Run decoding benchmark only"
+    )
+    args = parser.parse_args()
+
+    dtype_map = {
+        "bfloat16": torch.bfloat16,
+        "float16": torch.float16,
+        "float32": torch.float32,
+    }
+
     config = AutoRegressiveLMConfig(
         vocab_size=10000,
         dim=1536,
@@ -227,23 +271,29 @@ if __name__ == "__main__":
         norm_eps=1e-5,
     )
 
-    benchmark = GenerationBenchmark(config)
+    benchmark = GenerationBenchmark(
+        config, device=args.device, dtype=dtype_map[args.dtype], cache_type=args.cache
+    )
 
     print("=" * 80)
-    print("Running AutoRegressiveLM Generation Benchmark (KVCache)")
+    print(
+        f"Running AutoRegressiveLM Benchmark (device={args.device}, dtype={args.dtype})"
+    )
     print("=" * 80)
 
-    prefill_result = benchmark.run_prefill_benchmark(
-        batch_size=4,
-        prompt_length=512,
-        num_trials=5,
-    )
-    print_benchmark_result(prefill_result)
+    if not args.decode_only:
+        prefill_result = benchmark.run_prefill_benchmark(
+            batch_size=args.batch_size,
+            prompt_length=args.prompt_length,
+            num_trials=args.num_trials,
+        )
+        print_benchmark_result(prefill_result)
 
-    gen_result = benchmark.run_decoding_benchmark(
-        batch_size=4,
-        prompt_length=512,
-        gen_length=128,
-        num_trials=5,
-    )
-    print_benchmark_result(gen_result)
+    if not args.prefill_only:
+        gen_result = benchmark.run_decoding_benchmark(
+            batch_size=args.batch_size,
+            prompt_length=args.prompt_length,
+            gen_length=args.gen_length,
+            num_trials=args.num_trials,
+        )
+        print_benchmark_result(gen_result)

scripts/tools/generate.py

@@ -1,5 +1,6 @@
 import argparse
 import json
+from typing import Optional
 
 import torch
 
@@ -17,7 +18,7 @@ def processor(
     top_p: float,
     question_key: str,
     response_key: str,
-    max_tokens: int,
+    max_tokens: Optional[int],
     batch_size: int,
 ):
     # Load model and tokenizer
@@ -72,7 +73,7 @@ def processor(
 
 
 if __name__ == "__main__":
-    parser = argparse.ArgumentParser(description="Run generate with a Khaosz model.")
+    parser = argparse.ArgumentParser(description="Batch generation from JSONL file.")
 
     parser.add_argument(
         "--param_path", type=str, required=True, help="Path to the model directory."
@@ -93,36 +94,42 @@ if __name__ == "__main__":
         "--question_key",
         type=str,
         default="question",
-        help="Key for the question in the input JSON.",
+        help="Key for the question in the input JSON (default: question).",
     )
     parser.add_argument(
         "--response_key",
         type=str,
         default="response",
-        help="Key for the response in the output JSON.",
+        help="Key for the response in the output JSON (default: response).",
     )
     parser.add_argument(
         "--temperature",
         type=float,
         default=0.60,
-        help="Temperature for generating responses.",
+        help="Temperature for generating responses (default: 0.60).",
     )
     parser.add_argument(
-        "--top_k", type=int, default=30, help="Top-k value for generating responses."
+        "--top_k",
+        type=int,
+        default=30,
+        help="Top-k value for generating responses (default: 30).",
     )
     parser.add_argument(
         "--top_p",
         type=float,
         default=0.95,
-        help="Top-p value for generating responses.",
+        help="Top-p value for generating responses (default: 0.95).",
     )
     parser.add_argument(
-        "--batch_size", type=int, default=1, help="Batch size for generating responses."
+        "--batch_size",
+        type=int,
+        default=1,
+        help="Batch size for generating responses (default: 1).",
     )
     parser.add_argument(
         "--max_tokens",
         type=int,
-        default=2048,
+        default=None,
         help="Maximum tokens to generate (default: model config max_len).",
     )
 

scripts/tools/preprocess.py

@@ -0,0 +1,38 @@
+"""CLI: JSONL → tokenized .h5/.bin via config-driven Pipeline."""
+
+import argparse
+
+from astrai.config.preprocess_config import PipelineConfig
+from astrai.preprocessing.pipeline import Pipeline
+
+
+def main():
+    parser = argparse.ArgumentParser(
+        description="Raw JSONL → tokenized .h5/.bin via config-driven Pipeline"
+    )
+    parser.add_argument(
+        "inputs", nargs="+", metavar="JSONL", help="One or more JSONL files"
+    )
+    parser.add_argument("--output_dir", "-o", required=True, help="Output directory")
+    parser.add_argument(
+        "--config", "-c", required=True, help="Path to pipeline config JSON"
+    )
+    parser.add_argument(
+        "--tokenizer_path",
+        default="params",
+        help="Path to tokenizer directory (default: params)",
+    )
+    args = parser.parse_args()
+
+    config = PipelineConfig.from_file(args.config)
+
+    Pipeline(
+        config=config,
+        input_paths=args.inputs,
+        output_dir=args.output_dir,
+        tokenizer_path=args.tokenizer_path,
+    ).run()
+
+
+if __name__ == "__main__":
+    main()

scripts/tools/train.py

@@ -1,16 +1,97 @@
 import argparse
 import os
 from functools import partial
+from typing import Any, Dict
 
 import torch
 import torch.optim as optim
+from torch import Tensor, nn
 
 from astrai.config import AutoRegressiveLMConfig, TrainConfig
 from astrai.dataset import DatasetFactory
 from astrai.model import AutoRegressiveLM
+from astrai.model.components.decoder_block import DecoderBlock
 from astrai.trainer import SchedulerFactory, Trainer
 
 
+class MuonMix(optim.Optimizer):
+    """Combined Muon (matrix) + AdamW (non-matrix) optimizer."""
+
+    def __init__(
+        self,
+        model: nn.Module,
+        lr: float = 3e-4,
+        weight_decay: float = 0.1,
+        momentum: float = 0.95,
+        nesterov: bool = True,
+        ns_steps: int = 5,
+        adjust_lr_fn: str = "match_rms_adamw",
+    ):
+        defaults = dict(
+            lr=lr,
+            weight_decay=weight_decay,
+            momentum=momentum,
+            nesterov=nesterov,
+            ns_steps=ns_steps,
+            adjust_lr_fn=adjust_lr_fn,
+        )
+        params = [p for p in model.parameters() if p.requires_grad]
+        super().__init__(params, defaults)
+
+        matrix_params: list[Tensor] = []
+        other_params: list[Tensor] = []
+        for name, param in model.named_parameters():
+            if not param.requires_grad:
+                continue
+            if (
+                param.dim() >= 2
+                and "norm" not in name
+                and "bias" not in name
+                and "embed" not in name
+                and "lm_head" not in name
+            ):
+                matrix_params.append(param)
+            else:
+                other_params.append(param)
+
+        self.muon = optim.Muon(
+            matrix_params,
+            lr=lr,
+            weight_decay=weight_decay,
+            momentum=momentum,
+            nesterov=nesterov,
+            ns_steps=ns_steps,
+            adjust_lr_fn=adjust_lr_fn,
+        )
+        self.adamw = optim.AdamW(
+            [{"params": other_params, "weight_decay": 0.0}],
+            lr=lr,
+            betas=(0.9, 0.95),
+            fused=True,
+        )
+
+        self.param_groups = [*self.muon.param_groups, *self.adamw.param_groups]
+
+    @torch.no_grad()
+    def step(self, closure=None):
+        self.muon.step(closure)
+        self.adamw.step(closure)
+
+    def zero_grad(self, set_to_none: bool = True):
+        self.muon.zero_grad(set_to_none)
+        self.adamw.zero_grad(set_to_none)
+
+    def state_dict(self) -> Dict[str, Any]:
+        return {
+            "muon": self.muon.state_dict(),
+            "adamw": self.adamw.state_dict(),
+        }
+
+    def load_state_dict(self, state_dict: Dict[str, Any]):
+        self.muon.load_state_dict(state_dict["muon"])
+        self.adamw.load_state_dict(state_dict["adamw"])
+
+
 def parse_args() -> argparse.Namespace:
 
     parser = argparse.ArgumentParser(description="Train the AutoRegressiveLM model.")
@@ -63,22 +144,35 @@ def parse_args() -> argparse.Namespace:
         help="Max gradient norm for clipping.",
     )
     parser.add_argument(
-        "--adamw_beta1",
+        "--weight_decay",
         type=float,
-        default=0.9,
-        help="Beta1 for AdamW optimizer.",
+        default=0.1,
+        help="Weight decay (applied to Muon matrix params; non-matrix use 0).",
     )
     parser.add_argument(
-        "--adamw_beta2",
+        "--muon_momentum",
         type=float,
         default=0.95,
-        help="Beta2 for AdamW optimizer.",
+        help="Momentum factor for Muon optimizer.",
     )
     parser.add_argument(
-        "--adamw_weight_decay",
-        type=float,
-        default=0.01,
-        help="Weight decay for AdamW optimizer.",
+        "--muon_nesterov",
+        action=argparse.BooleanOptionalAction,
+        default=True,
+        help="Enable Nesterov momentum for Muon.",
+    )
+    parser.add_argument(
+        "--muon_ns_steps",
+        type=int,
+        default=5,
+        help="Newton-Schulz iteration steps for Muon.",
+    )
+    parser.add_argument(
+        "--muon_adjust_lr",
+        type=str,
+        default="match_rms_adamw",
+        choices=["original", "match_rms_adamw"],
+        help="Muon learning rate adjustment strategy.",
     )
     parser.add_argument(
         "--random_seed", type=int, default=3407, help="Random seed for reproducibility."
@@ -112,9 +206,15 @@ def parse_args() -> argparse.Namespace:
     parser.add_argument(
         "--label_smoothing",
         type=float,
-        default=0.05,
+        default=0.0,
         help="cross_entropy function label smoothing parameter",
     )
+    parser.add_argument(
+        "--gradient_checkpointing",
+        action=argparse.BooleanOptionalAction,
+        default=False,
+        help="Enable activation checkpointing for DecoderBlock modules.",
+    )
 
     parser.add_argument(
         "--ckpt_interval",
@@ -128,6 +228,30 @@ def parse_args() -> argparse.Namespace:
         default="checkpoint",
         help="Directory to save checkpoints.",
     )
+    parser.add_argument(
+        "--val_split",
+        type=float,
+        default=None,
+        help="Ratio to split from training dataset for validation (e.g. 0.05).",
+    )
+    parser.add_argument(
+        "--val_step",
+        type=int,
+        default=1000,
+        help="Number of optimizer steps between validation runs.",
+    )
+    parser.add_argument(
+        "--metrics",
+        nargs="*",
+        default=["loss", "lr", "grad_norm"],
+        help="Metrics to log (e.g. --metrics loss lr val_loss). Default: loss lr grad_norm.",
+    )
+    parser.add_argument(
+        "--log_dir",
+        type=str,
+        default="checkpoint/logs",
+        help="Directory for metric logs.",
+    )
     parser.add_argument(
         "--grpo_sync_interval",
         type=int,
@@ -138,9 +262,30 @@ def parse_args() -> argparse.Namespace:
         "--start_epoch", type=int, default=0, help="Start epoch for training."
     )
     parser.add_argument(
-        "--start_batch", type=int, default=0, help="Start batch for training."
+        "--start_samples",
+        type=int,
+        default=0,
+        help="Start samples (per rank) for training.",
     )
 
+    parser.add_argument(
+        "--master_addr",
+        type=str,
+        default="localhost",
+        help="Master node address for distributed training.",
+    )
+    parser.add_argument(
+        "--master_port",
+        type=str,
+        default="29500",
+        help="Master node port for distributed training.",
+    )
+    parser.add_argument(
+        "--backend",
+        type=str,
+        default="nccl",
+        help="Distributed training backend.",
+    )
     parser.add_argument("--nprocs", type=int, default=1, help="Number of GPUs to use.")
     parser.add_argument(
         "--parallel_mode",
@@ -159,6 +304,50 @@ def parse_args() -> argparse.Namespace:
         choices=["spawn", "fork", "forkserver"],
         help="Multiprocessing start method.",
     )
+    parser.add_argument(
+        "--neftune_alpha",
+        type=float,
+        default=0.0,
+        help="NEFTune noise alpha (0=disabled, typical: 5.0).",
+    )
+
+    parser.add_argument(
+        "--schedule_type",
+        type=str,
+        default="cosine",
+        choices=["cosine", "sgdr", "wsd"],
+        help="Learning rate scheduler type.",
+    )
+    parser.add_argument(
+        "--min_rate",
+        type=float,
+        default=None,
+        help="Minimum LR as fraction of base LR. Uses scheduler default if not set (cosine/sgdr: 0.05, wsd: 0.0).",
+    )
+    parser.add_argument(
+        "--cycle_length",
+        type=int,
+        default=None,
+        help="SGDR first cycle length in steps. Defaults to total_steps - warmup_steps.",
+    )
+    parser.add_argument(
+        "--t_mult",
+        type=int,
+        default=2,
+        help="SGDR cycle length multiplier per restart.",
+    )
+    parser.add_argument(
+        "--stable_steps",
+        type=int,
+        default=None,
+        help="WSD stable plateau steps. Required when --schedule_type wsd.",
+    )
+    parser.add_argument(
+        "--decay_steps",
+        type=int,
+        default=None,
+        help="WSD decay steps. Defaults to total_steps - warmup_steps - stable_steps.",
+    )
 
     args = parser.parse_args()
 
@@ -169,14 +358,15 @@ def create_model(config):
     return AutoRegressiveLM(config).to(dtype=torch.bfloat16)
 
 
-def create_optimizer(model, **kwargs) -> optim.Optimizer:
-    return optim.AdamW(model.parameters(), fused=True, **kwargs)
+def create_optimizer(model, **kwargs) -> MuonMix:
+    return MuonMix(model, **kwargs)
 
 
 def create_scheduler(
     optimizer: optim.Optimizer, **kwargs
 ) -> optim.lr_scheduler.LRScheduler:
-    return SchedulerFactory.create(optimizer, **kwargs)
+    schedule_type = kwargs.pop("schedule_type")
+    return SchedulerFactory.create(schedule_type, optimizer, **kwargs)
 
 
 def compute_total_steps(
@@ -200,34 +390,40 @@ def train(
     train_type: str,
     param_path: str,
     data_root_path: str,
-    max_lr: float,
     n_epoch: int,
     batch_per_device: int,
     start_epoch: int,
-    start_batch: int,
+    start_samples: int,
     grad_accum_steps: int,
     warmup_ratio: float,
     ckpt_interval: int,
     ckpt_dir: str,
-    dpo_beta: float,
-    grpo_clip_eps: float,
-    grpo_kl_coef: float,
-    group_size: int,
-    grpo_sync_interval: int,
-    adamw_beta1: float,
-    adamw_beta2: float,
-    adamw_weight_decay: float,
+    val_split: float,
+    val_step: int,
+    metrics: list[str],
+    log_dir: str,
     max_grad_norm: float,
-    label_smoothing: float,
     random_seed: int,
     num_workers: int,
     pin_memory: bool,
+    gradient_checkpointing: bool,
     window_size: int,
     stride: int,
     nprocs: int,
     parallel_mode: str,
     device_type: str,
+    backend: str,
+    master_addr: str,
+    master_port: str,
     start_method: str,
+    neftune_alpha: float,
+    schedule_type: str,
+    min_rate: float,
+    cycle_length: int,
+    t_mult: int,
+    stable_steps: int,
+    decay_steps: int,
+    **kwargs,
 ):
     assert train_type in ["seq", "sft", "dpo", "grpo"]
     assert os.path.exists(param_path)
@@ -237,17 +433,18 @@ def train(
     # Load config
     config_path = os.path.join(param_path, "config.json")
     config = AutoRegressiveLMConfig.from_file(config_path)
+    config.neftune_alpha = neftune_alpha
 
     if window_size is None:
         window_size = config.max_len
 
     strategy_kwargs = {
-        "beta": dpo_beta,
-        "label_smoothing": label_smoothing,
-        "clip_eps": grpo_clip_eps,
-        "kl_coef": grpo_kl_coef,
-        "group_size": group_size,
-        "sync_interval": grpo_sync_interval,
+        "beta": kwargs.pop("dpo_beta"),
+        "label_smoothing": kwargs.pop("label_smoothing"),
+        "clip_eps": kwargs.pop("grpo_clip_eps"),
+        "kl_coef": kwargs.pop("grpo_kl_coef"),
+        "group_size": kwargs.pop("group_size"),
+        "sync_interval": kwargs.pop("grpo_sync_interval"),
     }
 
     executor_kwargs = {
@@ -265,27 +462,46 @@ def train(
 
     optimizer_fn = partial(
         create_optimizer,
-        **{
-            "lr": max_lr,
-            "betas": (adamw_beta1, adamw_beta2),
-            "weight_decay": adamw_weight_decay,
-        },
+        lr=kwargs.pop("max_lr"),
+        weight_decay=kwargs.pop("weight_decay"),
+        momentum=kwargs.pop("muon_momentum"),
+        nesterov=kwargs.pop("muon_nesterov"),
+        ns_steps=kwargs.pop("muon_ns_steps"),
+        adjust_lr_fn=kwargs.pop("muon_adjust_lr"),
     )
 
     total_steps = compute_total_steps(
         len(dataset), n_epoch, batch_per_device, nprocs, grad_accum_steps
     )
     warmup_steps = int(warmup_ratio * total_steps)
+    warmup_steps = min(warmup_steps, total_steps)
+
+    scheduler_kwargs = {"warmup_steps": warmup_steps}
+
+    if schedule_type == "cosine":
+        scheduler_kwargs["lr_decay_steps"] = total_steps - warmup_steps
+    elif schedule_type == "sgdr":
+        scheduler_kwargs["cycle_length"] = cycle_length or (total_steps - warmup_steps)
+        scheduler_kwargs["t_mult"] = t_mult
+    elif schedule_type == "wsd":
+        remaining = total_steps - warmup_steps
+        stable_steps_ = stable_steps or max(1, int(remaining * 0.8))
+        scheduler_kwargs["stable_steps"] = stable_steps_
+        scheduler_kwargs["decay_steps"] = max(
+            1, decay_steps or (remaining - stable_steps_)
+        )
+
+    if min_rate is not None:
+        scheduler_kwargs["min_rate"] = min_rate
 
     scheduler_fn = partial(
         create_scheduler,
-        **{
-            "schedule_type": "cosine",
-            "warmup_steps": min(warmup_steps, total_steps),
-            "lr_decay_steps": total_steps - min(warmup_steps, total_steps),
-        },
+        schedule_type=schedule_type,
+        **scheduler_kwargs,
     )
 
+    grad_ckpt_modules = [DecoderBlock] if gradient_checkpointing else []
+
     train_config = TrainConfig(
         model_fn=model_fn,
         strategy=train_type,
@@ -296,7 +512,7 @@ def train(
         n_epoch=n_epoch,
         batch_per_device=batch_per_device,
         start_epoch=start_epoch,
-        start_batch=start_batch,
+        start_samples=start_samples,
         ckpt_interval=ckpt_interval,
         grad_accum_steps=grad_accum_steps,
         max_grad_norm=max_grad_norm,
@@ -304,11 +520,20 @@ def train(
         num_workers=num_workers,
         pin_memory=pin_memory,
         nprocs=nprocs,
+        backend=backend,
+        master_addr=master_addr,
+        master_port=master_port,
         parallel_mode=parallel_mode,
         device_type=device_type,
         start_method=start_method,
+        val_split=val_split,
+        val_step=val_step,
+        metrics=metrics,
+        log_dir=log_dir,
+        gradient_checkpointing_modules=grad_ckpt_modules,
         executor_kwargs=executor_kwargs,
         extra_kwargs=strategy_kwargs,
+        neftune_alpha=neftune_alpha,
     )
 
     trainer = Trainer(train_config)

setup.py

@@ -0,0 +1,61 @@
+import os
+import sys
+from pathlib import Path
+
+from setuptools import setup
+from setuptools.command.build_ext import build_ext as _build_ext
+
+sys.path.insert(0, str(Path(__file__).parent))
+os.makedirs("astrai/extension", exist_ok=True)
+
+
+def _should_build():
+    force = os.environ.get("CSRC_KERNELS", "").strip().lower()
+    if force == "true":
+        return True
+    if force == "false":
+        return False
+    try:
+        import shutil
+
+        import torch
+
+        return shutil.which("nvcc") is not None and torch.cuda.is_available()
+    except Exception:
+        return False
+
+
+ext_modules = []
+cmdclass = {}
+
+if _should_build():
+    import torch
+    from torch.utils.cpp_extension import BuildExtension, CUDAExtension
+
+    from csrc.build import REGISTRY
+
+    _torch_lib = torch.utils.cpp_extension.library_paths()[0]
+
+    for name, info in REGISTRY.items():
+        ext_modules.append(
+            CUDAExtension(
+                f"astrai.extension.{name}",
+                info["sources"],
+                extra_compile_args={
+                    "cxx": info["cxx_flags"],
+                    "nvcc": info["nvcc_flags"],
+                },
+                extra_link_args=[f"-Wl,-rpath,{_torch_lib}"],
+            )
+        )
+    cmdclass["build_ext"] = BuildExtension
+
+if not cmdclass:
+
+    class _NullBuildExt(_build_ext):
+        def build_extensions(self):
+            pass
+
+    cmdclass["build_ext"] = _NullBuildExt
+
+setup(ext_modules=ext_modules, cmdclass=cmdclass)

tests/conftest.py

@@ -75,7 +75,7 @@ class MultiTurnDataset(Dataset):
 
 
 class EarlyStoppingDataset(Dataset):
-    """Dataset that triggers early stopping after a specified number of iterations."""
+    """Dataset that triggers early stopping after consuming a specified number of samples."""
 
     def __init__(self, length=10, stop_after=5):
         self.length = length

tests/data/conftest.py

@@ -0,0 +1,249 @@
+import json
+import os
+import tempfile
+
+import pytest
+from tokenizers import Tokenizer, models, pre_tokenizers, trainers
+
+from astrai.config.preprocess_config import (
+    InputConfig,
+    PipelineConfig,
+    ProcessingConfig,
+)
+from astrai.preprocessing.builder import (
+    MultiOutputMaskBuilder,
+    SectionedMaskBuilder,
+    SingleOutputMaskBuilder,
+)
+from astrai.tokenize import AutoTokenizer
+
+_SPECIAL_TOKENS_CONFIG = {
+    "bos_token": "<|begin_of_sentence|>",
+    "eos_token": "<|end_of_sentence|>",
+    "pad_token": "<|_pad_|>",
+    "unk_token": "<|_unk_|>",
+    "im_start": "<|im_start|>",
+    "im_end": "<|im_end|>",
+}
+
+_SPECIAL_TOKENS = list(_SPECIAL_TOKENS_CONFIG.values())
+
+_CHAT_TEMPLATE = (
+    "{% for message in messages %}"
+    "{% if message['role'] == 'system' %}"
+    "<|im_start|>system\n{{ message['content'] }}<|im_end|>\n"
+    "{% elif message['role'] == 'user' %}"
+    "<|im_start|>user\n{{ message['content'] }}<|im_end|>\n"
+    "{% elif message['role'] == 'assistant' %}"
+    "<|im_start|>assistant\n{{ message['content'] }}<|im_end|>\n"
+    "{% endif %}"
+    "{% endfor %}"
+    "{% if add_generation_prompt %}<|im_start|>assistant\n{% endif %}"
+)
+
+_CHAT_SECTIONS = [{"field": "messages", "action": "$role", "template": True}]
+
+_INSTRUCTION_SECTIONS = [
+    {"field": "prompt", "action": "mask", "add_special_tokens": True},
+    {"field": "response", "action": "train"},
+]
+
+_TEXT_SECTIONS = [{"field": "text", "action": "train"}]
+
+_GRPO_RESPONSE_SECTIONS = [{"field": "responses", "action": "train"}]
+
+
+def _build_chat_tokenizer():
+    tok = Tokenizer(models.BPE())
+    tok.pre_tokenizer = pre_tokenizers.ByteLevel(add_prefix_space=False)
+    tr = trainers.BpeTrainer(
+        vocab_size=512,
+        min_frequency=1,
+        special_tokens=_SPECIAL_TOKENS,
+    )
+    train_data = [
+        "hello world",
+        "Hi there!",
+        "You are helpful.",
+        "What is 2+2?",
+        "Tell me a story about dragons and knights.",
+        "Sure, here is a tale.",
+        "Translate to French: Hello",
+        "Bonjour",
+        "Artificial Intelligence is a field of computer science.",
+        "system",
+        "user",
+        "assistant",
+        "<|im_start|>",
+        "<|im_end|>",
+        *[chr(i) for i in range(32, 127)],
+    ]
+    tok.train_from_iterator(train_data, tr)
+
+    auto_tok = AutoTokenizer()
+    auto_tok._tokenizer = tok
+    auto_tok._special_token_map = {
+        "bos_token": "<|begin_of_sentence|>",
+        "eos_token": "<|end_of_sentence|>",
+        "pad_token": "<|_pad_|>",
+        "unk_token": "<|_unk_|>",
+    }
+    auto_tok.set_chat_template(_CHAT_TEMPLATE)
+    return auto_tok
+
+
+@pytest.fixture(scope="session")
+def chat_tokenizer():
+    return _build_chat_tokenizer()
+
+
+@pytest.fixture
+def temp_dir():
+    d = tempfile.mkdtemp()
+    yield d
+    import shutil
+
+    shutil.rmtree(d, ignore_errors=True)
+
+
+def make_chat_config():
+    return PipelineConfig(
+        input=InputConfig(sections=_CHAT_SECTIONS),
+        mask={"system": "mask", "user": "mask", "assistant": "train"},
+        mask_default="mask",
+        preprocessing=ProcessingConfig(max_seq_len=2048),
+    )
+
+
+def make_instruction_config():
+    return PipelineConfig(
+        input=InputConfig(sections=_INSTRUCTION_SECTIONS),
+        mask={"prompt": "mask", "response": "train"},
+        mask_default="mask",
+        preprocessing=ProcessingConfig(max_seq_len=2048),
+    )
+
+
+def make_text_config():
+    return PipelineConfig(
+        input=InputConfig(sections=_TEXT_SECTIONS),
+        preprocessing=ProcessingConfig(
+            max_seq_len=2048, min_chars=1, max_chars=2_000_000
+        ),
+    )
+
+
+def make_dpo_chat_config():
+    return PipelineConfig(
+        input=InputConfig(
+            sources={
+                "chosen": {
+                    "sections": [
+                        {"field": "chosen", "action": "$role", "template": True}
+                    ]
+                },
+                "rejected": {
+                    "sections": [
+                        {"field": "rejected", "action": "$role", "template": True}
+                    ]
+                },
+            }
+        ),
+        mask={"user": "mask", "assistant": "train"},
+        mask_default="mask",
+        preprocessing=ProcessingConfig(max_seq_len=2048),
+    )
+
+
+def make_grpo_config():
+    return PipelineConfig(
+        input=InputConfig(
+            sources={
+                "prompts": {
+                    "sections": [
+                        {"field": "prompt", "action": "mask", "template": True}
+                    ]
+                },
+                "responses": {
+                    "sections": _GRPO_RESPONSE_SECTIONS,
+                    "list_field": True,
+                    "mask_key": "masks",
+                },
+                "rewards": {
+                    "sections": [{"field": "rewards", "action": "value"}],
+                },
+            }
+        ),
+        mask={"user": "mask", "assistant": "train"},
+        mask_default="mask",
+        preprocessing=ProcessingConfig(max_seq_len=2048),
+    )
+
+
+def make_grpo_no_template_config():
+    return PipelineConfig(
+        input=InputConfig(
+            sources={
+                "prompts": {
+                    "sections": [
+                        {
+                            "field": "prompt",
+                            "action": "mask",
+                            "add_special_tokens": True,
+                        }
+                    ]
+                },
+                "responses": {
+                    "sections": _GRPO_RESPONSE_SECTIONS,
+                    "list_field": True,
+                    "mask_key": "masks",
+                },
+                "rewards": {
+                    "sections": [{"field": "rewards", "action": "value"}],
+                },
+            }
+        ),
+        mask={"user": "mask", "assistant": "train"},
+        mask_default="mask",
+        preprocessing=ProcessingConfig(max_seq_len=2048),
+    )
+
+
+@pytest.fixture
+def builder():
+    return SectionedMaskBuilder()
+
+
+@pytest.fixture
+def single_builder():
+    return SingleOutputMaskBuilder()
+
+
+@pytest.fixture
+def multi_builder():
+    return MultiOutputMaskBuilder()
+
+
+@pytest.fixture
+def tokenizer_dir(temp_dir, test_tokenizer):
+    d = os.path.join(temp_dir, "tok")
+    os.makedirs(d, exist_ok=True)
+    test_tokenizer._tokenizer.save(os.path.join(d, "tokenizer.json"))
+    with open(os.path.join(d, "tokenizer_config.json"), "w") as f:
+        json.dump(
+            {"special_tokens": {"pad_token": "<|_pad_|>", "unk_token": "<|_unk_|>"}}, f
+        )
+    return d
+
+
+@pytest.fixture
+def chat_tokenizer_dir(temp_dir, chat_tokenizer):
+    d = os.path.join(temp_dir, "tok")
+    os.makedirs(d, exist_ok=True)
+    chat_tokenizer._tokenizer.save(os.path.join(d, "tokenizer.json"))
+    with open(os.path.join(d, "tokenizer_config.json"), "w") as f:
+        json.dump(
+            {"special_tokens": _SPECIAL_TOKENS_CONFIG, "chat_template": _CHAT_TEMPLATE},
+            f,
+        )
+    return d