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

[refactor] remove redundant code in linear

Browse Source
This commit is contained in:
Li Wei
2025-12-24 12:02:09 +08:00
parent 75d0bdae2f
commit 383eb5459a
8 changed files with 37 additions and 1570 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
vllm_kunlun/models/qwen3_omni_moe_thinker.py
View File

@@ -92,7 +92,7 @@ from vllm.multimodal.processing import (
from vllm.platforms.interface import _Backend
from vllm.sequence import IntermediateTensors
from vllm_kunlun.ops.linear import (
from vllm.model_executor.layers.linear import (
ColumnParallelLinear,
MergedColumnParallelLinear,
QKVParallelLinear,