diff --git a/vllm/model_executor/models/llama4.py b/vllm/model_executor/models/llama4.py
index e5d1a671f5d6..0fdc30f36f9b 100644
--- a/vllm/model_executor/models/llama4.py
+++ b/vllm/model_executor/models/llama4.py
@@ -273,8 +273,8 @@ def __init__(
             cache_config=cache_config,
             prefix=f"{prefix}.self_attn",
         )
-        is_moe_layer = (self.layer_idx +
-                        1) % config.interleave_moe_layer_step == 0
+        is_moe_layer = config.interleave_moe_layer_step > 0 and (
+            self.layer_idx + 1) % config.interleave_moe_layer_step == 0
         if is_moe_layer:
             self.feed_forward = Llama4MoE(
                 config=config,