phoenix/litellm
1
0
Fork 0
forked from phoenix/litellm-mirror
Code Pull requests Activity

fix(utils.py): support cache logging for async router calls

Browse source
This commit is contained in:
Krrish Dholakia 2023-12-13 19:11:43 -08:00
parent cffd190887
commit 73e5b96d8e
2 changed files with 52 additions and 3 deletions
Download patch file Download diff file Expand all files Collapse all files

2
litellm/utils.py
View file

@ -1660,7 +1660,7 @@ def client(original_function):
end_time = datetime.datetime.now()
model, custom_llm_provider, dynamic_api_key, api_base = litellm.get_llm_provider(model=model, custom_llm_provider=kwargs.get('custom_llm_provider', None), api_base=kwargs.get('api_base', None), api_key=kwargs.get('api_key', None))
print_verbose(f"Async Wrapper: Completed Call, calling async_success_handler: {logging_obj.async_success_handler}")
logging_obj.update_environment_variables(model=model, user=kwargs.get('user', None), optional_params={}, litellm_params={"logger_fn": kwargs.get('logger_fn', None), "acompletion": True}, input=kwargs.get('messages', ""), api_key=kwargs.get('api_key', None), original_response=str(cached_result), additional_args=None, stream=kwargs.get('stream', False))
logging_obj.update_environment_variables(model=model, user=kwargs.get('user', None), optional_params={}, litellm_params={"logger_fn": kwargs.get('logger_fn', None), "acompletion": True, "metadata": kwargs.get("metadata", {}), "model_info": kwargs.get("model_info", {}), "proxy_server_request": kwargs.get("proxy_server_request", None), "preset_cache_key": kwargs.get("preset_cache_key", None), "stream_response": kwargs.get("stream_response", {})}, input=kwargs.get('messages', ""), api_key=kwargs.get('api_key', None), original_response=str(cached_result), additional_args=None, stream=kwargs.get('stream', False))
asyncio.create_task(logging_obj.async_success_handler(cached_result, start_time, end_time, cache_hit))
threading.Thread(target=logging_obj.success_handler, args=(cached_result, start_time, end_time, cache_hit)).start()
return cached_result