Skip to content

GitLab

Unverified Commit b5815c84 authored by Woosuk Kwon's avatar Woosuk Kwon Committed by GitHub
Browse files

[V1] Fix non-cudagraph op name (#10166) 


Signed-off-by: default avatarWoosuk Kwon <woosuk.kwon@berkeley.edu>
parent 6b304715
Hide whitespace changes
Inline Side-by-side
Showing with 1 addition and 1 deletion
vllm/v1/worker/gpu_model_runner.py
View file @ b5815c84
...@@ -411,7 +411,7 @@ class GPUModelRunner: ...@@ -411,7 +411,7 @@ class GPUModelRunner:
set_compilation_config( set_compilation_config(
CompilationConfig( CompilationConfig(
use_cudagraph=True, use_cudagraph=True,
non_cudagraph_ops=["vllm.unified_flash_attention"], non_cudagraph_ops=["vllm.unified_v1_flash_attention"],
use_inductor=True, use_inductor=True,
)) ))
......
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment