Merge pull request #25728 from BerriAI/litellm_fix_together_ai_test_model

[Fix] Test - Together AI: replace deprecated Mixtral with serverless Qwen3.5-9B

tests/llm_translation/test_together_ai.py

@@ -20,7 +20,7 @@ import pytest
 class TestTogetherAI(BaseLLMChatTest):
     def get_base_completion_call_args(self) -> dict:
         litellm.set_verbose = True
-        return {"model": "together_ai/mistralai/Mixtral-8x7B-Instruct-v0.1"}
+        return {"model": "together_ai/Qwen/Qwen3.5-9B"}
 
     def test_tool_call_no_arguments(self, tool_call_no_arguments):
         """Test that tool calls with no arguments is translated correctly. Relevant issue: https://github.com/BerriAI/litellm/issues/6833"""

tests/local_testing/test_completion.py

@@ -65,7 +65,7 @@ def test_completion_custom_provider_model_name():
     try:
         litellm.cache = None
         response = completion(
-            model="together_ai/mistralai/Mixtral-8x7B-Instruct-v0.1",
+            model="together_ai/Qwen/Qwen3.5-9B",
             messages=messages,
             logger_fn=logger_fn,
         )
@@ -2815,7 +2815,7 @@ def test_customprompt_together_ai():
         print(litellm.success_callback)
         print(litellm._async_success_callback)
         response = completion(
-            model="together_ai/mistralai/Mixtral-8x7B-Instruct-v0.1",
+            model="together_ai/Qwen/Qwen3.5-9B",
             messages=messages,
             roles={
                 "system": {
@@ -3682,7 +3682,7 @@ def test_completion_together_ai_stream():
     messages = [{"content": user_message, "role": "user"}]
     try:
         response = completion(
-            model="together_ai/mistralai/Mixtral-8x7B-Instruct-v0.1",
+            model="together_ai/Qwen/Qwen3.5-9B",
             messages=messages,
             stream=True,
             max_tokens=5,

tests/local_testing/test_multiple_deployments.py

@@ -25,7 +25,7 @@ model_list = [
     {
         "model_name": "mistral-7b-instruct",
         "litellm_params": {  # params for litellm completion/embedding call
-            "model": "together_ai/mistralai/Mixtral-8x7B-Instruct-v0.1",
+            "model": "together_ai/Qwen/Qwen3.5-9B",
             "api_key": os.getenv("TOGETHERAI_API_KEY"),
         },
     },

tests/local_testing/test_text_completion.py

@@ -4034,7 +4034,7 @@ def test_async_text_completion_together_ai():
     async def test_get_response():
         try:
             response = await litellm.atext_completion(
-                model="together_ai/mistralai/Mixtral-8x7B-Instruct-v0.1",
+                model="together_ai/Qwen/Qwen3.5-9B",
                 prompt="good morning",
                 max_tokens=10,
             )