opencode.ai "mimo-v2.5" is unable to transmit and comprehend audio files

Description

• testcase1:
  -- request
  POST https://opencode.ai/zen/go/v1/chat/completions
  { "model": "mimo-v2.5", "messages": [ { "role": "system", "content": "你是一个宠物音频大师." }, { "role": "user", "content": [ { "type": "input_audio", "input_audio": { "data": "https://www.bryantmshop.site/voices/cat-crying.mp3" } }, { "type": "text", "text": "请描述内容" } ] } ], "stream": false }

response1:
{ "id": "gen-1785404018-jxvdMT0QxDzoNmqEdi3P", "object": "chat.completion", "created": 1785404018, "model": "mimo-v2.5", "choices": [ { "index": 0, "finish_reason": "stop", "logprobs": null, "message": { "role": "assistant", "content": "好的！请提供需要我描述的**音频内容**（例如：猫叫声、狗吠声、鸟鸣声、仓鼠的吱吱声等），并告诉我：\n\n1. **音频来源：** 是录音文件？视频片段？还是您现场听到的声音？\n2. **具体需求：** 您希望我如何描述？\n * 声音特征（音调、节奏、响度、音色）？\n * 情绪/状态推测（开心、焦虑、饥饿、警戒、求偶、疼痛）？\n * 可能的行为意图（要食物、要出门、求关注、警告）？\n * 物种/品种的大致判断？\n * 声音的背景环境？\n * 其他特殊分析需求？\n\n**请将音频内容或更详细的描述提供给我，我将竭尽全力为您进行专业、细致的宠物音频分析！** 🐾", "refusal": null, "reasoning": "嗯，用户只提供了“请描述内容”这四个字，

• testcase2
• request
  { "model": "mimo-v2.5", "messages": [ { "role": "system", "content": "你是一个宠物音频大师." }, { "role": "user", "content": [ { "type": "input_audio", "input_audio": { "data": "data:audio/mpeg;base64,$BASE64_AUDIO" } }, { "type": "text", "text": "请描述内容" } ] } ], "stream": false }
• response
  { "id": "gen-1785403625-Gn21AjYDUpdBrty9KEo2", "object": "chat.completion", "created": 1785403625, "model": "mimo-v2.5", "choices": [ { "index": 0, "finish_reason": "stop", "logprobs": null, "message": { "role": "assistant", "content": "您好！您发送的消息中似乎没有包含需要我描述的内容。\n\n请您提供具体的内容，例如：\n\n- **一段音频/视频**（请上传或描述）\n- **一张图片**（请上传）\n- **一段文字**（请粘贴）\n- **一个主题或话题**\n\n这样我就可以为您进行详细的描述了！😊", "refusal": null, "reasoning": "The user says \"请描述内容\" which means \"Please describe the content.\" However, there's no actual content provided for me to describe. The message appears to be empty or missing the content they want me to describe.\n\nLet me respond and ask them to provide the content they'd like me to describe.", "reasoning_details": [ { "type": "reasoning.text", "text": "The user says \"请描述内容\" which means \"Please describe the content.\" However, there's no actual content provided for me to describe. The message appears to be empty or missing the content they want me to describe.\n\nLet me respond and ask them to provide the content they'd like me to describe.", "format": "unknown", "index": 0 } ] } } ], "usage": { "prompt_tokens": 20, "completion_tokens": 135, "total_tokens": 155, "prompt_tokens_details": { "audio_tokens": 0, "cached_tokens": 0, "video_tokens": 0, "cache_write_tokens": 0 }, "completion_tokens_details": { "audio_tokens": 0, "reasoning_tokens": 62, "image_tokens": 0 } }, "cost": "0" }

Conclusion:opencode.ai "mimo-v2.5" is unable to transmit and comprehend audio files

references：https://mimo.mi.com/docs/zh-CN/quick-start/usage-guide/multimodal-understanding/audio-understanding?target=%E9%9F%B3%E9%A2%91%E4%BC%A0%E5%85%A5%E6%96%B9%E5%BC%8F

Plugins

No response

OpenCode version

版本1.18.3 (1.18.3)
I used the direct request server method

Steps to reproduce

1. post curl command to terminal

Screenshot and/or share link

Operating System

mac 14.7.5 (23H527)

Terminal

git bash