xc-llm-ascend

Files

Levi 4e728f1f40 [Bugfix] fix qwen3-vl-moe shape ERROR during the _prepare_inputs phase under high concurrency. (#4658 )

### What this PR does / why we need it?
Earlier we fixed a similar issue for qwen2.5-vl 【
https://github.com/vllm-project/vllm-ascend/issues/4430 】, and then the
multimodal models in vllm v0.11.0 should all have this problem. Here, we
have specifically proposed a fix for qwen3-vl-moe.

---------

Signed-off-by: Levi-JQ <yujinqi2@huawei.com>
Co-authored-by: Levi-JQ <yujinqi2@huawei.com>

2025-12-08 19:30:16 +08:00

layers

Revert "[Feat] Shared expert dp for deepseek and deepseek_mtp (#3495 )" (#3586 )

2025-10-21 22:24:30 +08:00

__init__.py

[0.11.0][Perf] Add padding vision tower for Qwen2_5_Omni (#4041 )

2025-11-08 13:56:05 +08:00

deepseek_v3_2.py

[DeepSeek] Seperate deepseek v3.2 modeling form deepseek v2 (#3531 )

2025-10-20 09:50:44 +08:00

qwen2_5_omni_thinker.py

[0.11.0][Perf] Add padding vision tower for Qwen2_5_Omni (#4041 )