test(ci): refresh retired OpenAI tool-call models (#43676)

tests/local_testing/test_alangfuse.py

@@ -690,7 +690,7 @@ def test_langfuse_logging_tool_calling():
     ]
 
     response = litellm.completion(
-        model="gpt-3.5-turbo-1106",
+        model="gpt-6-luna",
         messages=messages,
         tools=tools,
         tool_choice="auto",  # auto is default, but we'll be explicit
@@ -698,6 +698,8 @@ def test_langfuse_logging_tool_calling():
     print("\nLLM Response1:\n", response)
     response_message = response.choices[0].message
     tool_calls = response.choices[0].message.tool_calls
+    assert response.choices[0].message.tool_calls
+    assert all(call.function.name == "get_current_weather" for call in response.choices[0].message.tool_calls)
 
 
 # test_langfuse_logging_tool_calling()

tests/local_testing/test_function_calling.py

@@ -39,7 +39,7 @@ def get_current_weather(location, unit="fahrenheit"):
 @pytest.mark.parametrize(
     "model",
     [
-        "gpt-3.5-turbo-1106",
+        "gpt-6-luna",
         "mistral/mistral-large-latest",
         "claude-haiku-4-5-20251001",
         "gemini/gemini-2.5-flash-lite",
@@ -386,7 +386,7 @@ def test_parallel_function_call_stream():
             }
         ]
         response = litellm.completion(
-            model="gpt-3.5-turbo-1106",
+            model="gpt-6-luna",
             messages=messages,
             tools=tools,
             stream=True,
@@ -435,7 +435,7 @@ def test_parallel_function_call_stream():
                 )  # extend conversation with function response
             print(f"messages: {messages}")
             second_response = litellm.completion(
-                model="gpt-3.5-turbo-1106", messages=messages, temperature=0.2, seed=22
+                model="gpt-6-luna", messages=messages, temperature=0.2, seed=22, reasoning_effort="none"
             )  # get a new response from the model where it can see the function response
             print("second response\n", second_response)
             return second_response

tests/local_testing/test_lunary.py

@@ -83,13 +83,15 @@ def test_lunary_with_tools():
     ]
 
     response = litellm.completion(
-        model="gpt-3.5-turbo-1106",
+        model="gpt-6-luna",
         messages=messages,
         tools=tools,
         tool_choice="auto",  # auto is default, but we'll be explicit
     )
 
     response_message = response.choices[0].message
+    assert response.choices[0].message.tool_calls
+    assert all(call.function.name == "get_current_weather" for call in response.choices[0].message.tool_calls)
     print("\nLLM Response:\n", response.choices[0].message)