0z5a a144d7f306 perf: accelerate decode linear with bf16 gemv ...

- add decode-shape benchmark harness
- add bf16 GEMV CUDA primitive with head-dim generic kernel
- dispatch decode-time linear layers to gemv for M=1
- extend gemv coverage to small decode batches

2026-09-02 13:11:25 +08:00

..

demo

refactor: centralize logging in astrai.logging, replace ASTRAI_TIMED with log level

2026-08-08 12:39:27 +08:00

docker

refactor: move runtime parsers into scripts/docker

2026-08-27 12:33:32 +08:00

eval

perf: increase eval batch sizes and add max_seq_len

2026-07-30 23:55:37 +08:00

tools

perf: accelerate decode linear with bf16 gemv

2026-09-02 13:11:25 +08:00

docker.sh

feat: rework docker workflow for gpu-first training

2026-08-13 22:51:13 +08:00

pre_commit.sh

ci: 优化 GitHub Actions 工作流

2026-04-05 22:40:16 +08:00

serve.sh

refactor: move runtime parsers into scripts/docker

2026-08-27 12:33:32 +08:00

train.sh

refactor: move runtime parsers into scripts/docker

2026-08-27 12:33:32 +08:00