diff --git a/test-quality-budget.json b/test-quality-budget.json
index 5039143eaa0..38fd31d7275 100644
--- a/test-quality-budget.json
+++ b/test-quality-budget.json
@@ -12,7 +12,7 @@
     "limit": 469
   },
   "TQ005": {
-    "limit": 2459
+    "limit": 2436
   },
   "TQ006": {
     "limit": 34
diff --git a/tests/test_litellm/llms/azure/test_azure_common_utils.py b/tests/test_litellm/llms/azure/test_azure_common_utils.py
index 99826c14069..3cc251b6228 100644
--- a/tests/test_litellm/llms/azure/test_azure_common_utils.py
+++ b/tests/test_litellm/llms/azure/test_azure_common_utils.py
@@ -812,7 +812,6 @@ async def test_azure_client_reuse(function_name, is_async, args):
     """
     Test that multiple Azure API calls reuse the same Azure OpenAI client
     """
-    litellm.set_verbose = True
 
     # Determine which client class to mock based on whether the test is async
     client_path = (
diff --git a/tests/test_litellm/llms/bedrock/embed/test_bedrock_async_invoke_embedding.py b/tests/test_litellm/llms/bedrock/embed/test_bedrock_async_invoke_embedding.py
index 8b6034d1133..1c802ecd077 100644
--- a/tests/test_litellm/llms/bedrock/embed/test_bedrock_async_invoke_embedding.py
+++ b/tests/test_litellm/llms/bedrock/embed/test_bedrock_async_invoke_embedding.py
@@ -153,7 +153,6 @@ class TestBedrockAsyncInvokeEmbedding:
 
     def test_async_invoke_twelvelabs_embedding_with_mock(self):
         """Test async invoke embedding with mocked HTTP calls."""
-        litellm.set_verbose = True
         client = HTTPHandler()
         test_api_key = "test-bearer-token-12345"
         model = "bedrock/async_invoke/twelvelabs.marengo-embed-2-7-v1:0"
@@ -193,7 +192,6 @@ class TestBedrockAsyncInvokeEmbedding:
     @pytest.mark.asyncio
     async def test_async_invoke_twelvelabs_embedding_async_with_mock(self):
         """Test async invoke embedding with async calls."""
-        litellm.set_verbose = True
         client = AsyncHTTPHandler()
         test_api_key = "test-bearer-token-12345"
         model = "bedrock/async_invoke/twelvelabs.marengo-embed-2-7-v1:0"
diff --git a/tests/test_litellm/llms/bedrock/image/test_bedrock_image_bearer_token.py b/tests/test_litellm/llms/bedrock/image/test_bedrock_image_bearer_token.py
index 41ac030ff07..b2b00d25051 100644
--- a/tests/test_litellm/llms/bedrock/image/test_bedrock_image_bearer_token.py
+++ b/tests/test_litellm/llms/bedrock/image/test_bedrock_image_bearer_token.py
@@ -18,7 +18,6 @@ mock_image_response = {"images": ["base64_encoded_image_data"], "error": None}
 class TestBedrockImageGeneration:
     def test_image_generation_with_api_key_bearer_token(self):
         """Test image generation with bearer token authentication"""
-        litellm.set_verbose = True
         test_api_key = "test-bearer-token-12345"
         model = "bedrock/stability.sd3-large-v1:0"
         prompt = "A cute baby sea otter"
@@ -53,7 +52,6 @@ class TestBedrockImageGeneration:
 
     def test_image_generation_with_env_variable_bearer_token(self, monkeypatch):
         """Test image generation with bearer token from environment variable"""
-        litellm.set_verbose = True
         test_api_key = "env-bearer-token-12345"
         model = "bedrock/stability.sd3-large-v1:0"
         prompt = "A cute baby sea otter"
@@ -90,7 +88,6 @@ class TestBedrockImageGeneration:
     @pytest.mark.asyncio
     async def test_async_image_generation_with_bearer_token(self):
         """Test async image generation with bearer token authentication"""
-        litellm.set_verbose = True
         test_api_key = "async-bearer-token-12345"
         model = "bedrock/stability.sd3-large-v1:0"
         prompt = "A cute baby sea otter"
@@ -125,7 +122,6 @@ class TestBedrockImageGeneration:
 
     def test_image_generation_with_sigv4(self):
         """Test image generation falls back to SigV4 auth when no bearer token is provided"""
-        litellm.set_verbose = True
         model = "bedrock/stability.sd3-large-v1:0"
         prompt = "A cute baby sea otter"
 
diff --git a/tests/test_litellm/llms/bedrock/rerank/test_bedrock_rerank_header_forwarding.py b/tests/test_litellm/llms/bedrock/rerank/test_bedrock_rerank_header_forwarding.py
index 17443ca899e..d8259652641 100644
--- a/tests/test_litellm/llms/bedrock/rerank/test_bedrock_rerank_header_forwarding.py
+++ b/tests/test_litellm/llms/bedrock/rerank/test_bedrock_rerank_header_forwarding.py
@@ -66,7 +66,6 @@ def test_bedrock_rerank_header_forwarding_sync(model):
     This test verifies the fix for the issue where headers configured via
     forward_client_headers_to_llm_api were not being passed to Bedrock rerank provider.
     """
-    litellm.set_verbose = True
     client = HTTPHandler()
     test_api_key = "test-bearer-token-12345"
 
@@ -160,7 +159,6 @@ async def test_bedrock_rerank_header_forwarding_async(model):
     This test verifies the fix for the issue where headers configured via
     forward_client_headers_to_llm_api were not being passed to Bedrock rerank provider.
     """
-    litellm.set_verbose = True
     client = AsyncHTTPHandler()
     test_api_key = "test-bearer-token-12345"
 
@@ -332,7 +330,6 @@ def test_bedrock_rerank_extra_headers_and_headers_merge():
     This ensures that headers from kwargs (forwarded by proxy) and extra_headers
     (passed explicitly) are both included in the final headers sent to the provider.
     """
-    litellm.set_verbose = True
     client = HTTPHandler()
     test_api_key = "test-bearer-token-12345"
     model = "bedrock/arn:aws:bedrock:us-east-1::foundation-model/cohere.rerank-v3-5:0"
diff --git a/tests/test_litellm/llms/openai/test_openai_common_utils.py b/tests/test_litellm/llms/openai/test_openai_common_utils.py
index bfd681cc06e..bef8d02b0df 100644
--- a/tests/test_litellm/llms/openai/test_openai_common_utils.py
+++ b/tests/test_litellm/llms/openai/test_openai_common_utils.py
@@ -86,7 +86,6 @@ async def test_openai_client_reuse(function_name, is_async, args):
     """
     Test that multiple API calls reuse the same OpenAI client
     """
-    litellm.set_verbose = True
 
     # Determine which client class to mock based on whether the test is async
     client_path = (
diff --git a/tests/test_litellm/llms/vertex_ai/test_vertex.py b/tests/test_litellm/llms/vertex_ai/test_vertex.py
index ec73e5e42be..ae260a2d887 100644
--- a/tests/test_litellm/llms/vertex_ai/test_vertex.py
+++ b/tests/test_litellm/llms/vertex_ai/test_vertex.py
@@ -33,7 +33,6 @@ def test_completion_pydantic_obj_2():
 
     from litellm.llms.custom_httpx.http_handler import HTTPHandler
 
-    litellm.set_verbose = True
 
     class CalendarEvent(BaseModel):
         name: str
@@ -259,7 +258,6 @@ def test_vertex_tool_type_field_removal():
 def test_function_calling_with_gemini():
     from litellm.llms.custom_httpx.http_handler import HTTPHandler
 
-    litellm.set_verbose = True
     client = HTTPHandler()
     with patch.object(client, "post", new=MagicMock()) as mock_post:
         try:
@@ -310,7 +308,6 @@ def test_function_calling_with_gemini():
 
 
 def test_multiple_function_call():
-    litellm.set_verbose = True
     from litellm.llms.custom_httpx.http_handler import HTTPHandler
 
     client = HTTPHandler()
@@ -420,7 +417,6 @@ def test_multiple_function_call():
 
 
 def test_multiple_function_call_changed_text_pos():
-    litellm.set_verbose = True
     from litellm.llms.custom_httpx.http_handler import HTTPHandler
 
     client = HTTPHandler()
@@ -528,7 +524,6 @@ def test_multiple_function_call_changed_text_pos():
 
 
 def test_function_calling_with_gemini_multiple_results():
-    litellm.set_verbose = True
     from litellm.llms.custom_httpx.http_handler import HTTPHandler
 
     client = HTTPHandler()
@@ -1103,7 +1098,6 @@ def test_logprobs_unit_test():
 
 
 def test_logprobs():
-    litellm.set_verbose = True
     from litellm.llms.custom_httpx.http_handler import HTTPHandler
 
     client = HTTPHandler()
diff --git a/tests/test_litellm/llms/volcengine/test_volcengine_embedding.py b/tests/test_litellm/llms/volcengine/test_volcengine_embedding.py
index 1670dac0e9d..04caecab478 100644
--- a/tests/test_litellm/llms/volcengine/test_volcengine_embedding.py
+++ b/tests/test_litellm/llms/volcengine/test_volcengine_embedding.py
@@ -31,7 +31,6 @@ class TestVolcEngineEmbedding(BaseLLMEmbeddingTest):
     @pytest.mark.parametrize("sync_mode", [True, False])
     async def test_basic_embedding(self, sync_mode):
         """Test basic embedding functionality with realistic response"""
-        litellm.set_verbose = True
         embedding_call_args = self.get_base_embedding_call_args()
 
         # Mock the embedding functions to avoid actual API calls
diff --git a/tests/test_litellm/proxy/guardrails/guardrail_hooks/test_cato_networks.py b/tests/test_litellm/proxy/guardrails/guardrail_hooks/test_cato_networks.py
index c23fbc0234e..caed64ef417 100644
--- a/tests/test_litellm/proxy/guardrails/guardrail_hooks/test_cato_networks.py
+++ b/tests/test_litellm/proxy/guardrails/guardrail_hooks/test_cato_networks.py
@@ -26,7 +26,6 @@ from litellm.proxy.guardrails.init_guardrails import init_guardrails_v2
 
 
 def test_cato_guard_config():
-    litellm.set_verbose = True
     litellm.guardrail_name_config_map = {}
 
     init_guardrails_v2(
@@ -47,7 +46,6 @@ def test_cato_guard_config():
 
 def test_cato_guard_config_no_api_key(monkeypatch):
     monkeypatch.delenv("CATO_API_KEY", raising=False)
-    litellm.set_verbose = True
     litellm.guardrail_name_config_map = {}
     with pytest.raises(CatoNetworksGuardrailMissingSecrets, match="Couldn't get Cato Networks api key"):
         init_guardrails_v2(
diff --git a/tests/test_litellm/proxy/guardrails/guardrail_hooks/test_cisco_ai_defense_chat.py b/tests/test_litellm/proxy/guardrails/guardrail_hooks/test_cisco_ai_defense_chat.py
index 8974a18593b..779075a40d9 100644
--- a/tests/test_litellm/proxy/guardrails/guardrail_hooks/test_cisco_ai_defense_chat.py
+++ b/tests/test_litellm/proxy/guardrails/guardrail_hooks/test_cisco_ai_defense_chat.py
@@ -44,7 +44,6 @@ from tests.test_litellm.proxy.guardrails.guardrail_hooks._cisco_ai_defense_test_
 
 def test_cisco_ai_defense_config_via_init_v2_chat(monkeypatch):
     monkeypatch.setenv("CISCO_AI_DEFENSE_API_KEY", "test-key")
-    litellm.set_verbose = True
     litellm.guardrail_name_config_map = {}
 
     init_guardrails_v2(
diff --git a/tests/test_litellm/proxy/guardrails/test_pillar_guardrails.py b/tests/test_litellm/proxy/guardrails/test_pillar_guardrails.py
index 48f6b3ba2b9..02123bc8c76 100644
--- a/tests/test_litellm/proxy/guardrails/test_pillar_guardrails.py
+++ b/tests/test_litellm/proxy/guardrails/test_pillar_guardrails.py
@@ -65,7 +65,6 @@ def setup_and_teardown():
     asyncio.set_event_loop(loop)
 
     # Set up litellm state
-    litellm.set_verbose = True
     litellm.guardrail_name_config_map = {}
 
     yield
diff --git a/tests/test_litellm/responses/test_text_format_conversion.py b/tests/test_litellm/responses/test_text_format_conversion.py
index a48540b129b..339b73c2729 100644
--- a/tests/test_litellm/responses/test_text_format_conversion.py
+++ b/tests/test_litellm/responses/test_text_format_conversion.py
@@ -158,7 +158,6 @@ class TestTextFormatConversion:
             new=mock_handler,
         ):
             litellm._turn_on_debug()
-            litellm.set_verbose = True
 
             # Call aresponses with text_format parameter
             response = await litellm.aresponses(