zhink d89b6dd43f adapter qwen3 moe attr for init (#3066) ...

adapter qwen3 moe attr for init

2025-07-30 16:49:28 +08:00

..

graph_optimization

update flake8 version to support pre-commit in python3.12 (#3000)

2025-07-24 01:43:31 -07:00

guided_decoding

Unify server-side and model-side Config (Part3) (#3047)

2025-07-29 17:07:44 +08:00

layers

w4a8 offline (#3074)

2025-07-30 16:33:30 +08:00

models

adapter qwen3 moe attr for init (#3066)

2025-07-30 16:49:28 +08:00

ops

[Feature] Support repetition early stop (#3024)

2025-07-29 22:42:54 +08:00

__init__.py

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

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

forward_meta.py

use dist.all_reduce(min) to sync num_blocks_local (#2933)

2025-07-21 01:23:36 -07:00

load_weight_utils.py

[fix] w4a8 model loading and hadamard config (#3013)

2025-07-28 18:17:59 +08:00

model_loader.py

Unify server-side and model-side Config (Part3) (#3047)

2025-07-29 17:07:44 +08:00

pre_and_post_process.py

[stop sequence] support stop sequence (#3025)

2025-07-29 14:17:37 +08:00