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

[BugFix] Fix Qwen2.5_Omni vision customized op attr err (#4568)

Browse Source 
Qwen2.5_Omni vision tower use AscendRMSNorm, which conatins a property
function. It would be override by set_forward_context(), patch
Qwen2_5OmniThinkerForConditionalGeneration func with customized
_process_image_input() and _process_video_input() to fix it.

### What this PR does / why we need it?

Fix Qwen2.5_Omni model infer image/video issue

### Does this PR introduce _any_ user-facing change?

No

### How was this patch tested?

- vLLM version: v0.11.2
- vLLM main: https://github.com/vllm-project/vllm/commit/v0.11.2

Signed-off-by: Ting FU <futing10@huawei.com>
This commit is contained in:
Ting FU
2025-12-01 09:18:55 +08:00
committed by GitHub
parent c68ddc11ce
commit e8e20c0bbf
2 changed files with 73 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

@@ -28,4 +28,5 @@ 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_qwen2_5_vl # noqa
import vllm_ascend.patch.worker.patch_qwen2_5_omni # noqa
import vllm_ascend.patch.worker.patch_rope # noqa