diff --git a/vllm/model_executor/layers/quantization/kernels/scaled_mm/pytorch.py b/vllm/model_executor/layers/quantization/kernels/scaled_mm/pytorch.py
index 7c4c64215a8e9..40e55cc97392c 100644
--- a/vllm/model_executor/layers/quantization/kernels/scaled_mm/pytorch.py
+++ b/vllm/model_executor/layers/quantization/kernels/scaled_mm/pytorch.py
@@ -230,9 +230,6 @@ class RowWiseTorchScaledMMLinearKernel(TorchScaledMMLinearKernel):
 
 
 class ChannelWiseTorchScaledMMLinearKernel(TorchScaledMMLinearKernel):
-    @classmethod
-    def get_min_capability(cls) -> int:
-        return 94
 
     @classmethod
     def can_implement(cls, c: FP8ScaledMMLinearLayerConfig) -> tuple[bool, str | None]: