From a9c6156137f89e45b6572d8bd652f721ce7bd309 Mon Sep 17 00:00:00 2001 From: Yuneng Jiang Date: Tue, 14 Apr 2026 17:31:33 -0700 Subject: [PATCH] [Fix] Test - Together AI: replace deprecated Mixtral with serverless Qwen3.5-9B Mixtral-8x7B-Instruct-v0.1 is no longer on Together AI's serverless tier and now requires a dedicated endpoint, causing multiple tests to fail in CI: - test_together_ai.py::TestTogetherAI::test_empty_tools - test_completion.py::test_completion_together_ai_stream - test_completion.py::test_customprompt_together_ai - test_completion.py::test_completion_custom_provider_model_name - test_text_completion.py::test_async_text_completion_together_ai Qwen/Qwen3.5-9B is currently serverless on Together AI and supports function calling, satisfying BaseLLMChatTest capability requirements. --- tests/llm_translation/test_together_ai.py | 2 +- tests/local_testing/test_completion.py | 6 +++--- tests/local_testing/test_multiple_deployments.py | 2 +- tests/local_testing/test_text_completion.py | 2 +- 4 files changed, 6 insertions(+), 6 deletions(-) diff --git a/tests/llm_translation/test_together_ai.py b/tests/llm_translation/test_together_ai.py index 023b7cfa77..5225ab78f6 100644 --- a/tests/llm_translation/test_together_ai.py +++ b/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""" diff --git a/tests/local_testing/test_completion.py b/tests/local_testing/test_completion.py index ef34c9f85b..f18a2b4afb 100644 --- a/tests/local_testing/test_completion.py +++ b/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, diff --git a/tests/local_testing/test_multiple_deployments.py b/tests/local_testing/test_multiple_deployments.py index 1c34cc5745..61baa73da0 100644 --- a/tests/local_testing/test_multiple_deployments.py +++ b/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"), }, }, diff --git a/tests/local_testing/test_text_completion.py b/tests/local_testing/test_text_completion.py index ab2153af8d..dde5f67ea1 100644 --- a/tests/local_testing/test_text_completion.py +++ b/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, )