周周周 17b414c2df MoE Default use triton's blockwise fp8 in TP Case (#3678)

2025-08-29 11:07:30 +08:00

..

best_practices

MoE Default use triton's blockwise fp8 in TP Case (#3678)

2025-08-29 11:07:30 +08:00

features

Revert "[Feature] block sparse attention (#3209)" (#3647)

2025-08-27 17:35:04 +08:00

get_started

[MetaxGPU] adapt to the latest fastdeploy on metax gpu (#3492)

2025-08-25 17:44:20 +08:00

online_serving

[Feature] bad words support v1 scheduler and specifiy token ids (#3608)

2025-08-25 20:14:51 -07:00

quantization

polish code with new pre-commit rule (#2923)

2025-07-19 23:19:27 +08:00

usage

MoE Default use triton's blockwise fp8 in TP Case (#3678)

2025-08-29 11:07:30 +08:00

zh

MoE Default use triton's blockwise fp8 in TP Case (#3678)

2025-08-29 11:07:30 +08:00

benchmark.md

polish code with new pre-commit rule (#2923)

2025-07-19 23:19:27 +08:00

index.md

Update README (#3426)

2025-08-15 18:46:28 +08:00

offline_inference.md

rename ernie_xxx to ernie4_5_xxx (#3621)

2025-08-26 19:29:27 +08:00

parameters.md

[Precision] Support lm_head layer running in float32 (#3597)

2025-08-27 11:34:53 +08:00

requirements.txt

Sync v2.0 version of code to github repo

2025-06-29 23:29:37 +00:00

supported_models.md

[Feature] multi source download (#3005)

2025-07-24 17:42:09 +08:00