diff --git a/vllm/model_executor/layers/quantization/compressed_tensors/compressed_tensors_moe.py b/vllm/model_executor/layers/quantization/compressed_tensors/compressed_tensors_moe.py
index 8504ba73defb..de63b9817612 100644
--- a/vllm/model_executor/layers/quantization/compressed_tensors/compressed_tensors_moe.py
+++ b/vllm/model_executor/layers/quantization/compressed_tensors/compressed_tensors_moe.py
@@ -542,11 +542,12 @@ def __init__(
         self.rocm_aiter_moe_enabled = is_rocm_aiter_moe_enabled()
 
         # cutlass path
+        self.is_fp8_w8a8_sm90 = quant_config._is_fp8_w8a8_sm90(
+            self.weight_quant, self.input_quant)
         self.is_fp8_w8a8_sm100 = quant_config._is_fp8_w8a8_sm100(
             self.weight_quant, self.input_quant)
-        self.use_cutlass = not self.block_quant and (
-            quant_config._is_fp8_w8a8_sm90(self.weight_quant, self.input_quant)
-            or self.is_fp8_w8a8_sm100)
+        self.use_cutlass = not self.block_quant and (self.is_fp8_w8a8_sm90
+                                            or self.is_fp8_w8a8_sm100)
         self.disable_expert_map = False
 
     def create_weights(self, layer: torch.nn.Module, num_experts: int,
@@ -1013,8 +1014,9 @@ def apply(
         elif self.use_cutlass:
             assert self.moe_quant_config is not None
 
-            # small-batch fallback on SM100
-            if self.is_fp8_w8a8_sm100 and topk_ids.shape[0] <= 8:
+            # SM90 or small-batch fallback on SM100
+            if self.is_fp8_w8a8_sm90 or (self.is_fp8_w8a8_sm100
+                                        and topk_ids.shape[0] <= 8):
                 from vllm.model_executor.layers.fused_moe import fused_experts
                 assert per_act_token == per_channel_quant
                 return fused_experts(