xinyun/vllm
1
0
Fork 0
mirror of https://git.datalinker.icu/vllm-project/vllm.git synced 2026-05-28 01:27:04 +08:00
Code Issues Packages Projects Releases Wiki Activity

Squelch MLA warning for Compressed-Tensors Models (#12704)

Browse Source 
Signed-off-by: Kyle Sayers <kylesayrs@gmail.com>
This commit is contained in:
Kyle Sayers 2025-02-03 16:29:56 -05:00 committed by GitHub
parent c11de33dad
commit 6dd5e52823
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 4 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

6
vllm/config.py
View File

@ -986,6 +986,9 @@ class ModelConfig:
@property @property
def use_mla(self) -> bool: def use_mla(self) -> bool:
if not self.is_deepseek_mla or envs.VLLM_MLA_DISABLE:
return False
if self.quantization is not None and self.quantization not in [\ if self.quantization is not None and self.quantization not in [\
"fp8", "compressed-tensors"]: "fp8", "compressed-tensors"]:
logger.warning( logger.warning(
@ -1012,8 +1015,7 @@ class ModelConfig:
quant_config) quant_config)
return False return False
use_mla = (self.is_deepseek_mla and not envs.VLLM_MLA_DISABLE) return True
return use_mla
@property @property
def supported_runner_types(self) -> Set[RunnerType]: def supported_runner_types(self) -> Set[RunnerType]: