xinyun/vllm
1
0
Fork 0
mirror of https://git.datalinker.icu/vllm-project/vllm.git synced 2025-12-19 06:35:01 +08:00
Code Issues Packages Projects Releases Wiki Activity

[Kernel] changing fused moe kernel chunk size default to 32k (#7995)

Browse Source
This commit is contained in:
Avshalom Manevich 2024-08-30 11:11:39 +07:00 committed by GitHub
parent 80c7b089b1
commit 34a0e96d46
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 1 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
vllm/envs.py
View File

@ -352,7 +352,7 @@ environment_variables: Dict[str, Callable[[], Any]] = {
os.path.join(get_default_cache_root(), "vllm", "xla_cache"), os.path.join(get_default_cache_root(), "vllm", "xla_cache"),
)), )),
"VLLM_FUSED_MOE_CHUNK_SIZE": "VLLM_FUSED_MOE_CHUNK_SIZE":
lambda: int(os.getenv("VLLM_FUSED_MOE_CHUNK_SIZE", "65536")), lambda: int(os.getenv("VLLM_FUSED_MOE_CHUNK_SIZE", "32768")),
# If set, vllm will skip the deprecation warnings. # If set, vllm will skip the deprecation warnings.
"VLLM_NO_DEPRECATION_WARNING": "VLLM_NO_DEPRECATION_WARNING":