diff --git a/test-quality-budget.json b/test-quality-budget.json
index 17baf64601b..fcb29c3191d 100644
--- a/test-quality-budget.json
+++ b/test-quality-budget.json
@@ -12,7 +12,7 @@
     "limit": 770
   },
   "TQ005": {
-    "limit": 2835
+    "limit": 2832
   },
   "TQ006": {
     "limit": 34
diff --git a/tests/image_gen_tests/test_image_generation.py b/tests/image_gen_tests/test_image_generation.py
index 33dcdbb57a5..ad141a651e8 100644
--- a/tests/image_gen_tests/test_image_generation.py
+++ b/tests/image_gen_tests/test_image_generation.py
@@ -105,20 +105,6 @@ def load_vertex_ai_credentials():
     os.environ["GOOGLE_APPLICATION_CREDENTIALS"] = os.path.abspath(temp_file.name)
 
 
-class TestVertexImageGeneration(BaseImageGenTest):
-    def get_base_image_generation_call_args(self) -> dict:
-        # comment this when running locally
-        load_vertex_ai_credentials()
-
-        litellm.in_memory_llm_clients_cache = InMemoryCache()
-        return {
-            "model": "vertex_ai/imagen-3.0-fast-generate-001",
-            "vertex_ai_project": "litellm-ci-cd",
-            "vertex_ai_location": "us-central1",
-            "n": 1,
-        }
-
-
 class TestVertexAIGeminiImageGeneration(BaseImageGenTest):
     """Test Gemini image generation models (Nano Banana)"""
 
diff --git a/tests/llm_translation/test_openai.py b/tests/llm_translation/test_openai.py
index 1a14b00c7d7..61819dfc860 100644
--- a/tests/llm_translation/test_openai.py
+++ b/tests/llm_translation/test_openai.py
@@ -422,7 +422,7 @@ def test_openai_web_search():
     """Makes a simple web search request and validates the response contains web search annotations and all expected fields are present"""
     litellm._turn_on_debug()
     response = litellm.completion(
-        model="openai/gpt-4o-search-preview",
+        model="openai/gpt-5-search-api",
         messages=[
             {
                 "role": "user",
@@ -442,7 +442,7 @@ def test_openai_web_search_streaming():
     # litellm._turn_on_debug()
     test_openai_web_search: Optional[ChatCompletionAnnotation] = None
     response = litellm.completion(
-        model="openai/gpt-4o-search-preview",
+        model="openai/gpt-5-search-api",
         messages=[
             {
                 "role": "user",
diff --git a/tests/logging_callback_tests/test_built_in_tools_cost_tracking.py b/tests/logging_callback_tests/test_built_in_tools_cost_tracking.py
index 335661d46d0..0e73ad834da 100644
--- a/tests/logging_callback_tests/test_built_in_tools_cost_tracking.py
+++ b/tests/logging_callback_tests/test_built_in_tools_cost_tracking.py
@@ -105,7 +105,7 @@ async def test_openai_web_search_logging_cost_tracking(
     from litellm._uuid import uuid
 
     request_kwargs = {
-        "model": "openai/gpt-4o-search-preview",
+        "model": "openai/gpt-5-search-api",
         "messages": [
             {
                 "role": "user",
diff --git a/tests/test_openai_endpoints.py b/tests/test_openai_endpoints.py
index 0d44064997e..ab43d1acb00 100644
--- a/tests/test_openai_endpoints.py
+++ b/tests/test_openai_endpoints.py
@@ -550,7 +550,7 @@ async def test_proxy_all_models():
     async with aiohttp.ClientSession() as session:
         # call chat/completions with a model that the key was not created for + the model is not on the config.yaml
         await chat_completion(
-            session=session, key=LITELLM_MASTER_KEY, model="groq/llama-3.1-8b-instant"
+            session=session, key=LITELLM_MASTER_KEY, model="groq/openai/gpt-oss-120b"
         )
 
         await chat_completion(