mimo-v2.5 request must chat_template_kwargs bug
general openai format request:
}'
{"id":"chatcmpl-4a319121-cdfd-4309-aded-0d6982c260e4","object":"chat.completion","created":1777450538,"model":"mimo-v2.5","choices":[{"index":0,"message":{"role":"assistant","content":null,"refusal":null,"annotations":null,"audio":null,"function_call":null,"tool_calls":[],"reasoning":"Hello! How can I help you today?"},"logprobs":null,"finish_reason":"stop","stop_reason":null,"token_ids":null}],"service_tier":null,"system_fingerprint":"vllm-0.1.dev16124+g4d7f212b2-tp4-ep-f9d51453","usage":{"prompt_tokens":27,"total_tokens":37,"completion_tokens":10,"prompt_tokens_details":null,"time_to_first_token_ms":156,"time_per_output_token_ms":0,"tokens_per_second":0},"prompt_logprobs":null,"prompt_token_ids":null,"kv_transfer_params":null}
response error into reasoning , context it null ,this is not very normal like other llm
but ,if i must bring "chat_template_kwargs": {"enable_thinking": true} ,the response it correct
{"id":"chatcmpl-2a8820fd-a8e8-4293-903a-abc608656e47","object":"chat.completion","created":1777450888,"model":"mimo-v2.5","choices":[{"index":0,"message":{"role":"assistant","content":"Hello there! 👋 I'm MiMo, nice to meet you! How can I help you today? Whether you have questions, need assistance with something, or just want to chat, I'm here for you! 😊","refusal":null,"annotations":null,"audio":null,"function_call":null,"tool_calls":[],"reasoning":"The user is greeting me with "Hello MiMo!" - they're addressing me by my name. Let me respond in a friendly and welcoming manner."},"logprobs":null,"finish_reason":"stop","stop_reason":null,"token_ids":null}],"service_tier":null,"system_fingerprint":"vllm-0.1.dev16124+g4d7f212b2-tp4-ep-f9d51453","usage":{"prompt_tokens":28,"total_tokens":107,"completion_tokens":79,"prompt_tokens_details":null,"time_per_output_token_ms":0,"tokens_per_second":0,"time_to_first_token_ms":413},"prompt_logprobs":null,"prompt_token_ids":null,"kv_transfer_params":null}
is this a bug vllm or mimo self?
mimo model it a thinking model or no thinking model?
mimo-v2.5 request must chat_template_kwargs bug
general openai format request:
}'
{"id":"chatcmpl-4a319121-cdfd-4309-aded-0d6982c260e4","object":"chat.completion","created":1777450538,"model":"mimo-v2.5","choices":[{"index":0,"message":{"role":"assistant","content":null,"refusal":null,"annotations":null,"audio":null,"function_call":null,"tool_calls":[],"reasoning":"Hello! How can I help you today?"},"logprobs":null,"finish_reason":"stop","stop_reason":null,"token_ids":null}],"service_tier":null,"system_fingerprint":"vllm-0.1.dev16124+g4d7f212b2-tp4-ep-f9d51453","usage":{"prompt_tokens":27,"total_tokens":37,"completion_tokens":10,"prompt_tokens_details":null,"time_to_first_token_ms":156,"time_per_output_token_ms":0,"tokens_per_second":0},"prompt_logprobs":null,"prompt_token_ids":null,"kv_transfer_params":null}
response error into reasoning , context it null ,this is not very normal like other llm
but ,if i must bring "chat_template_kwargs": {"enable_thinking": true} ,the response it correct
{"id":"chatcmpl-2a8820fd-a8e8-4293-903a-abc608656e47","object":"chat.completion","created":1777450888,"model":"mimo-v2.5","choices":[{"index":0,"message":{"role":"assistant","content":"Hello there! 👋 I'm MiMo, nice to meet you! How can I help you today? Whether you have questions, need assistance with something, or just want to chat, I'm here for you! 😊","refusal":null,"annotations":null,"audio":null,"function_call":null,"tool_calls":[],"reasoning":"The user is greeting me with "Hello MiMo!" - they're addressing me by my name. Let me respond in a friendly and welcoming manner."},"logprobs":null,"finish_reason":"stop","stop_reason":null,"token_ids":null}],"service_tier":null,"system_fingerprint":"vllm-0.1.dev16124+g4d7f212b2-tp4-ep-f9d51453","usage":{"prompt_tokens":28,"total_tokens":107,"completion_tokens":79,"prompt_tokens_details":null,"time_per_output_token_ms":0,"tokens_per_second":0,"time_to_first_token_ms":413},"prompt_logprobs":null,"prompt_token_ids":null,"kv_transfer_params":null}
is this a bug vllm or mimo self?
mimo model it a thinking model or no thinking model?