diff --git a/vllm/model_executor/layers/quantization/modelopt.py b/vllm/model_executor/layers/quantization/modelopt.py index 72178a6e6d..6566d67153 100644 --- a/vllm/model_executor/layers/quantization/modelopt.py +++ b/vllm/model_executor/layers/quantization/modelopt.py @@ -250,7 +250,7 @@ class ModelOptQuantConfigBase(QuantizationConfig): """ if hf_quant_cfg is None: return None - if hf_quant_cfg.get("quant_method", "").lower() != "modelopt": + if not hf_quant_cfg.get("quant_method", "").lower().startswith("modelopt"): return None if "quantization" in hf_quant_cfg: quant_config = hf_quant_cfg["quantization"]