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

[bugfix] fix quant method validation bug (#4831)

Browse Source 
### What this PR does / why we need it?
When `hf_quant_cfg` is not None and `hf_quant_cfg.quant_method == ""`,
func `override_quantization_method` will return None and raise
ValidationError.

### Does this PR introduce _any_ user-facing change?

### How was this patch tested?

- vLLM version: v0.12.0
- vLLM main:
ad32e3e19c

Signed-off-by: zzzzwwjj <1183291235@qq.com>
This commit is contained in:
zzzzwwjj
2025-12-09 23:42:01 +08:00
committed by GitHub
parent 863a5a5a17
commit f404c9af7f
2 changed files with 9 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
vllm_ascend/quantization/quant_config.py
View File

@@ -96,7 +96,7 @@ class AscendQuantConfig(QuantizationConfig):
user_quant) -> Optional[str]:
if hf_quant_cfg is not None:
quant_method = hf_quant_cfg.get("quant_method", None)
if quant_method is None and torch.npu.is_available():
if not quant_method and torch.npu.is_available():
return ASCEND_QUANTIZATION_METHOD
return None