xc-llm-ascend

Files

whx 72695c97d0 [BugFix][main] Fix quantization related mtp bug with patch (#3620 )

vLLM 0.11.0 didn't bring PR
(https://github.com/vllm-project/vllm/pull/25805) thus missing the
prefix of mtp's SharedHead. This PR fixes this bug with a patch to
vllm's deepseek_mtp. main also need this bugfix to support vllm's
v0.11.0

- vLLM version: v0.11.0rc3
- vLLM main: https://github.com/vllm-project/vllm/commit/v0.11.0

---------

Signed-off-by: whx-sjtu <2952154980@qq.com>

2025-10-23 09:54:31 +08:00

__init__.py

[BugFix][main] Fix quantization related mtp bug with patch (#3620 )

2025-10-23 09:54:31 +08:00

patch_deepseek_mtp.py

[BugFix][main] Fix quantization related mtp bug with patch (#3620 )

2025-10-23 09:54:31 +08:00

patch_distributed.py

[Refactor] refactor patch module (#3555 )

2025-10-21 20:19:46 +08:00

patch_logits.py

[Refactor] refactor patch module (#3555 )