diff --git a/vllm/model_executor/models/llama.py b/vllm/model_executor/models/llama.py
index 23141124e69e1..f43a40a0bfd34 100644
--- a/vllm/model_executor/models/llama.py
+++ b/vllm/model_executor/models/llama.py
@@ -57,7 +57,7 @@ class LlamaMLP(nn.Module):
         hidden_size: int,
         intermediate_size: int,
         hidden_act: str,
-        quant_config: Optional[QKVParallelLinear] = None,
+        quant_config: Optional[QuantizationConfig] = None,
         bias: bool = False,
     ) -> None:
         super().__init__()