biondizzle/vllm
1
0
Fork 0
Code Issues Pull Requests Actions 2 Packages Projects Releases Wiki Activity

[Misc] Do not print async output warning for v1 (#21151)

Browse Source 
Signed-off-by: Woosuk Kwon <woosuk.kwon@berkeley.edu>
This commit is contained in:
Woosuk Kwon
2025-07-17 21:57:02 -07:00
committed by GitHub
parent 5780121c95
commit 54cf1cae62
2 changed files with 2 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
vllm/platforms/cuda.py
View File

@@ -99,7 +99,7 @@ class CudaPlatformBase(Platform):
@classmethod
def is_async_output_supported(cls, enforce_eager: Optional[bool]) -> bool:
if enforce_eager:
if enforce_eager and not envs.VLLM_USE_V1:
logger.warning(
"To see benefits of async output processing, enable CUDA "
"graph. Since, enforce-eager is enabled, async output "

2
vllm/platforms/rocm.py
View File

@@ -299,7 +299,7 @@ class RocmPlatform(Platform):
@classmethod
def is_async_output_supported(cls, enforce_eager: Optional[bool]) -> bool:
if enforce_eager:
if enforce_eager and not envs.VLLM_USE_V1:
logger.warning(
"To see benefits of async output processing, enable CUDA "
"graph. Since, enforce-eager is enabled, async output "