Commit Graph

4 Commits

Author SHA1 Message Date
ViperEkura e8e228d035 feat: add optional CUDA kernel system (csrc/) + fused GQA decode attention
Structure:
  csrc/               -- .cu sources + build.py registry
  astrai/extension/   -- compiled .so + __init__.py (import dispatcher)
  setup.py            -- CUDAExtension from csrc/build.py REGISTRY

Control: CSRC_KERNELS=true|false env var at install time.
Fallback: astrai.extension.available dict for runtime detection.
2026-07-06 12:09:58 +08:00
ViperEkura a33d086883 build: 修改build 方式 2026-02-27 17:52:28 +08:00
ViperEkura 7a21f5d72e build(setup): 更新版本号并调整 Python 版本要求 2025-11-09 16:40:20 +08:00
ViperEkura a4443765ee Initial commit 2025-09-27 12:02:22 +08:00