-
Notifications
You must be signed in to change notification settings - Fork 19
Description
the function calling response of qwen/qwen-2.5-72b-instruct has stop reason "stop".
i try gpt-4o-mini, gemini-flash-2.0 and qwen-turbo/plux/max, their function callings all have stop reason "tool_calls".
why the inconsistency?
ChatCompletion(id='gen-1742353600-TuyQwrw32JlcRhYG2t96', choices=[Choice(finish_reason='stop', index=0, logprobs=None, message=ChatCompletionMessage(content='', refusal=None, role='assistant', annotations=None, audio=None, function_call=None, tool_calls=[ChatCompletionMessageToolCall(id='call_Li7QiC9LXUTiwc0X13e3JZ5Z', function=Function(arguments='{"search_terms": ["James", "Joyce"]}', name='search_gutenberg_books'), type='function', index=0)]), native_finish_reason='stop')], created=1742353600, model='qwen/qwen-2.5-72b-instruct', object='chat.completion', service_tier=None, system_fingerprint=None, usage=CompletionUsage(completion_tokens=28, prompt_tokens=481, total_tokens=509, completion_tokens_details=None, prompt_tokens_details=None), provider='DeepInfra')