diff --git a/litellm/llms/vertex_ai/videos/transformation.py b/litellm/llms/vertex_ai/videos/transformation.py
index 9389a5f419d..bac3d31501b 100644
--- a/litellm/llms/vertex_ai/videos/transformation.py
+++ b/litellm/llms/vertex_ai/videos/transformation.py
@@ -12,6 +12,7 @@ from typing import TYPE_CHECKING, Any, Final, cast
 import httpx
 from httpx._types import RequestFiles
 
+import litellm
 from litellm.constants import DEFAULT_GOOGLE_VIDEO_DURATION_SECONDS
 from litellm.images.utils import ImageEditRequestUtils
 from litellm.llms.base_llm.videos.transformation import BaseVideoConfig
@@ -146,7 +147,7 @@ class VertexAIVideoConfig(BaseVideoConfig, VertexBase):
         - prompt → prompt (in instances)
         - input_reference → image (in instances)
         - size → aspectRatio (e.g., "1280x720" → "16:9")
-        - size → resolution for Veo 3 models when inferable
+        - size → resolution for models with resolution-tier pricing when inferable
           ("1280x720"/"720x1280" → "720p", "1920x1080"/"1080x1920" → "1080p");
           skipped if ``resolution`` is already set
         - seconds → durationSeconds (defaults to 4 seconds if not provided)
@@ -177,7 +178,7 @@ class VertexAIVideoConfig(BaseVideoConfig, VertexBase):
                 has_resolution = "resolution" in mapped_params or (
                     isinstance(nested_params, dict) and nested_params.get("resolution") is not None
                 )
-                supports_resolution = model.removeprefix("vertex_ai/").startswith("veo-3.")
+                supports_resolution = self._supports_resolution_inference(model)
                 if supports_resolution and not has_resolution:
                     inferred_resolution = self._convert_size_to_resolution(size)
                     if inferred_resolution is not None:
@@ -210,6 +211,12 @@ class VertexAIVideoConfig(BaseVideoConfig, VertexBase):
     def _convert_size_to_resolution(self, size: str) -> str | None:
         return self._OPENAI_VIDEO_SIZE_TO_RESOLUTION.get(size)
 
+    @staticmethod
+    def _supports_resolution_inference(model: str) -> bool:
+        model_key = model if model.startswith("vertex_ai/") else f"vertex_ai/{model}"
+        model_info = litellm.model_cost.get(model_key, {})
+        return model_info.get("output_cost_per_second_1080p") is not None
+
     def validate_environment(
         self,
         headers: dict,
diff --git a/tests/test_litellm/llms/vertex_ai/videos/test_vertex_video_transformation.py b/tests/test_litellm/llms/vertex_ai/videos/test_vertex_video_transformation.py
index 1c2952cbb0b..11cea7c3d23 100644
--- a/tests/test_litellm/llms/vertex_ai/videos/test_vertex_video_transformation.py
+++ b/tests/test_litellm/llms/vertex_ai/videos/test_vertex_video_transformation.py
@@ -288,15 +288,33 @@ class TestVertexAIVideoConfig:
         assert "resolution" not in mapped
 
     @pytest.mark.parametrize(
-        ("size", "expected_resolution"),
-        (("1280x720", "720p"), ("1920x1080", "1080p")),
+        ("model", "size", "expected_resolution"),
+        (
+            (VEO_31_LITE_VERTEX_MODEL, "1280x720", "720p"),
+            (
+                VEO_31_LITE_VERTEX_MODEL.removeprefix("vertex_ai/"),
+                "1920x1080",
+                "1080p",
+            ),
+        ),
     )
-    def test_map_openai_size_to_resolution_for_veo_3(
-        self, size: str, expected_resolution: str
+    def test_map_openai_size_to_resolution_for_resolution_tier_model(
+        self,
+        model: str,
+        size: str,
+        expected_resolution: str,
+        monkeypatch: pytest.MonkeyPatch,
     ):
+        model_cost = _load_model_cost_map(BACKUP_MODEL_COST_PATH)
+        monkeypatch.setitem(
+            litellm.model_cost,
+            VEO_31_LITE_VERTEX_MODEL,
+            dict(model_cost[VEO_31_LITE_VERTEX_MODEL]),
+        )
+
         mapped = self.config.map_openai_params(
             video_create_optional_params={"size": size},
-            model=VEO_31_LITE_VERTEX_MODEL,
+            model=model,
             drop_params=False,
         )
 
@@ -313,6 +331,23 @@ class TestVertexAIVideoConfig:
         assert mapped["aspectRatio"] == "16:9"
         assert "resolution" not in mapped
 
+    def test_map_openai_size_does_not_infer_resolution_for_existing_veo_3(
+        self, monkeypatch: pytest.MonkeyPatch
+    ):
+        model = "veo-3.1-generate-001"
+        model_key = f"vertex_ai/{model}"
+        model_cost = _load_model_cost_map(BACKUP_MODEL_COST_PATH)
+        monkeypatch.setitem(litellm.model_cost, model_key, dict(model_cost[model_key]))
+
+        mapped = self.config.map_openai_params(
+            video_create_optional_params={"size": "1920x1080"},
+            model=model,
+            drop_params=False,
+        )
+
+        assert mapped["aspectRatio"] == "16:9"
+        assert "resolution" not in mapped
+
     def test_map_openai_size_does_not_override_provider_resolution(self):
         mapped = self.config.map_openai_params(
             video_create_optional_params={