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

[Quickfix] Add the missing apply_router_weight_on_input in FusedMoE init (#2348)

Browse Source 
### What this PR does / why we need it?
Add the missing `apply_router_weight_on_input` in FusedMoE init
Quick fix on
https://github.com/vllm-project/vllm-ascend/pull/2268#discussion_r2265828849

### Does this PR introduce _any_ user-facing change?
N/A

### How was this patch tested?
CI passed with existing test.


- vLLM version: v0.10.0
- vLLM main:
6807af8f46

Signed-off-by: MengqingCao <cmq0113@163.com>
This commit is contained in:
Mengqing Cao
2025-08-14 09:17:50 +08:00
committed by GitHub
parent 0f7492d18e
commit 8914d5a4b2
1 changed files with 1 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

1
vllm_ascend/ops/fused_moe.py
View File

@@ -1258,6 +1258,7 @@ class AscendFusedMoE(FusedMoE):
scoring_func=scoring_func, scoring_func=scoring_func,
e_score_correction_bias=e_score_correction_bias, e_score_correction_bias=e_score_correction_bias,
activation=activation, activation=activation,
apply_router_weight_on_input=apply_router_weight_on_input,
) )
AscendFusedMoE.moe_counter += 1 AscendFusedMoE.moe_counter += 1
self.moe_instance_id = AscendFusedMoE.moe_counter self.moe_instance_id = AscendFusedMoE.moe_counter