diff --git a/vllm/model_executor/layers/quantization/modelopt.py b/vllm/model_executor/layers/quantization/modelopt.py index 70922b5e83..de0b0655ef 100644 --- a/vllm/model_executor/layers/quantization/modelopt.py +++ b/vllm/model_executor/layers/quantization/modelopt.py @@ -405,7 +405,7 @@ class ModelOptFp8Config(ModelOptQuantConfigBase): @classmethod def get_min_capability(cls) -> int: - return 89 + return 80 @classmethod def override_quantization_method(