diff --git a/vllm_mlx/engine/batched.py b/vllm_mlx/engine/batched.py
index d94cfbb9..3aeda361 100644
--- a/vllm_mlx/engine/batched.py
+++ b/vllm_mlx/engine/batched.py
@@ -23,13 +23,28 @@ class MLLMModelWrapper:
     BatchGenerator expects model output to be subscriptable (logits array),
     but MLLM models return LanguageModelOutput objects. This wrapper extracts
     the logits from the output.
+
+    Additionally, some models like Gemma 3 require `pixel_values` as a required
+    positional argument, while others like Qwen2-VL make it optional. This wrapper
+    ensures `pixel_values=None` is passed for text-only requests.
     """
 
     def __init__(self, model):
         self._model = model
 
     def __call__(self, *args, **kwargs):
-        """Call the model and extract logits from LanguageModelOutput."""
+        """Call the model and extract logits from LanguageModelOutput.
+
+        Handles both models where pixel_values is optional (Qwen2-VL) and
+        models where it's required (Gemma 3) by ensuring pixel_values=None
+        is passed for text-only requests.
+        """
+        # For text-only requests, BatchGenerator calls model(input_ids, cache=cache)
+        # But Gemma 3 requires pixel_values as 2nd positional arg.
+        # Inject pixel_values=None if not provided and only input_ids is passed
+        if 'pixel_values' not in kwargs and len(args) == 1:
+            kwargs['pixel_values'] = None
+
         output = self._model(*args, **kwargs)
         # If output has logits attribute, return just the logits
         if hasattr(output, 'logits'):