diff --git a/vllm/model_executor/models/qwen2_5_vl.py b/vllm/model_executor/models/qwen2_5_vl.py
index 9d42ace2c8e8e..998cefd33e801 100644
--- a/vllm/model_executor/models/qwen2_5_vl.py
+++ b/vllm/model_executor/models/qwen2_5_vl.py
@@ -1587,20 +1587,3 @@ class Qwen2_5_VLForConditionalGeneration(
         vision_config = hf_config.vision_config
         merge_size = vision_config.spatial_merge_size
         return num_vision_tokens // merge_size**2
-
-    def get_allowed_mm_limits(self) -> Mapping[str, int]:
-        """Return the maximum allowed number of items for each modality."""
-        supported_mm_limits = self.get_supported_mm_limits()
-        mm_config = self.ctx.get_mm_config()
-
-        allowed_limits = dict[str, int]()
-        for modality, supported_limit in supported_mm_limits.items():
-            user_limit = mm_config.get_limit_per_prompt(modality)
-
-            allowed_limits[modality] = (
-                user_limit
-                if supported_limit is None
-                else min(user_limit, supported_limit)
-            )
-
-        return allowed_limits
diff --git a/vllm/model_executor/models/qwen2_vl.py b/vllm/model_executor/models/qwen2_vl.py
index a13859a2a71c3..cd9ddaa532490 100644
--- a/vllm/model_executor/models/qwen2_vl.py
+++ b/vllm/model_executor/models/qwen2_vl.py
@@ -1491,23 +1491,6 @@ class Qwen2VLForConditionalGeneration(
             tower_model="visual.",
         )
 
-    def get_allowed_mm_limits(self) -> Mapping[str, int]:
-        """Return the maximum allowed number of items for each modality."""
-        supported_mm_limits = self.get_supported_mm_limits()
-        mm_config = self.ctx.get_mm_config()
-
-        allowed_limits = dict[str, int]()
-        for modality, supported_limit in supported_mm_limits.items():
-            user_limit = mm_config.get_limit_per_prompt(modality)
-
-            allowed_limits[modality] = (
-                user_limit
-                if supported_limit is None
-                else min(user_limit, supported_limit)
-            )
-
-        return allowed_limits
-    
     def get_num_mm_encoder_tokens(
         self,
         num_image_tokens: int,
diff --git a/vllm/model_executor/models/qwen3_vl.py b/vllm/model_executor/models/qwen3_vl.py
index 18c0fd68afdc4..be0e5f8759d17 100644
--- a/vllm/model_executor/models/qwen3_vl.py
+++ b/vllm/model_executor/models/qwen3_vl.py
@@ -2110,20 +2110,3 @@ class Qwen3VLForConditionalGeneration(
         vision_config = hf_config.vision_config
         merge_size = vision_config.spatial_merge_size
         return num_vision_tokens // merge_size**2
-
-    def get_allowed_mm_limits(self) -> Mapping[str, int]:
-        """Return the maximum allowed number of items for each modality."""
-        supported_mm_limits = self.get_supported_mm_limits()
-        mm_config = self.ctx.get_mm_config()
-
-        allowed_limits = dict[str, int]()
-        for modality, supported_limit in supported_mm_limits.items():
-            user_limit = mm_config.get_limit_per_prompt(modality)
-
-            allowed_limits[modality] = (
-                user_limit
-                if supported_limit is None
-                else min(user_limit, supported_limit)
-            )
-
-        return allowed_limits