diff --git a/vllm/model_executor/kernels/linear/scaled_mm/cutlass.py b/vllm/model_executor/kernels/linear/scaled_mm/cutlass.py
index 9f69ab0c7377..20e370cbc011 100644
--- a/vllm/model_executor/kernels/linear/scaled_mm/cutlass.py
+++ b/vllm/model_executor/kernels/linear/scaled_mm/cutlass.py
@@ -166,6 +166,10 @@ def is_supported(
     ) -> tuple[bool, str | None]:
         if not current_platform.is_cuda():
             return False, "requires CUDA."
+
+        if compute_capability is not None and compute_capability < 89:
+            return False, "requires compute capability 89 and above."
+
         return True, None
 
     @classmethod
diff --git a/vllm/model_executor/layers/quantization/modelopt.py b/vllm/model_executor/layers/quantization/modelopt.py
index 24ec55e40063..0ccc4f21d735 100644
--- a/vllm/model_executor/layers/quantization/modelopt.py
+++ b/vllm/model_executor/layers/quantization/modelopt.py
@@ -2299,13 +2299,13 @@ def get_supported_act_dtypes(self) -> list[torch.dtype]:
 
     @classmethod
     def get_min_capability(cls) -> int:
-        # Ampere (SM80/SM86): NVFP4 routed experts run via Marlin W4A16, and FP8
+        # Turing (SM75): NVFP4 routed experts run via Marlin W4A16, and FP8
         # weight-only dense layers run via MarlinFP8 (W8A16, compute in
         # bf16/fp16). FP8 MoE, if present, also routes to Marlin because
         # TritonExperts gates its FP8 schemes behind supports_fp8() (cc>=89).
-        # None of these paths require native FP8 tensor cores, so SM80 is
+        # None of these paths require native FP8 tensor cores, so SM75 is
         # sufficient.
-        return 80
+        return 75
 
     @classmethod
     def override_quantization_method(