EngineX/xc-llm-ascend
3
0
Fork 0
Code Issues Pull Requests Actions Projects Releases Wiki Activity

[BugFix][v0.11.0] Fix quantization related mtp bug with patch (#3619)

Browse Source 
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.

---------

Signed-off-by: whx-sjtu <2952154980@qq.com>
This commit is contained in:
whx
2025-10-22 23:06:09 +08:00
committed by GitHub
parent 6e72bfdc50
commit 6464c97ff9
3 changed files with 70 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

1
vllm_ascend/patch/worker/__init__.py
View File

@@ -27,3 +27,4 @@ import vllm_ascend.patch.worker.patch_roberta # noqa
import vllm_ascend.patch.worker.patch_weight_loader # noqa
import vllm_ascend.patch.worker.patch_multimodal_merge # noqa
import vllm_ascend.patch.worker.patch_minicpm # noqa
import vllm_ascend.patch.worker.patch_deepseek_mtp # noqa