diff --git a/vllm/multimodal/processing.py b/vllm/multimodal/processing.py
index dc5c4307ce48..8e3f32698a6b 100644
--- a/vllm/multimodal/processing.py
+++ b/vllm/multimodal/processing.py
@@ -1186,10 +1186,16 @@ def get_hf_processor(
 
             typ = ProcessorMixin
 
+        from vllm.tokenizers.mistral import MistralTokenizer
+
+        tokenizer = self.tokenizer
+        if isinstance(tokenizer, MistralTokenizer):
+            tokenizer = tokenizer.transformers_tokenizer
+
         return cached_processor_from_config(
             self.model_config,
             processor_cls=typ,
-            tokenizer=self.tokenizer,
+            tokenizer=tokenizer,
             **kwargs,
         )