MiMO 报错 `The reasoning_content in the thinking mode must be passed back to the API.`

RT，佬们在使用 MiMO v2.5 的时候会有上面的报错吗？好像是晚上大规模出现，猜测可能有什么升级？

脱敏后的请求如下

{
  "model": "mimo-v2.5",
  "temperature": 0.6,
  "user": "69ef64ec580a2a75b3d5df9b",
  "stream": true,
  "tools": [
    {
      "type": "function",
      "function": {
        "name": "query_structured",
        "description": "Query",
        "parameters": {
          "type": "object",
          "properties": {
            "query": {
              "title": "Query",
              "type": "string"
            }
          },
          "required": [
            "query"
          ],
          "title": "query_structuredArguments"
        }
      }
    },
    {
      "type": "function",
      "function": {
        "name": "image_gen_oai",
        "description": "Generates high-quality, original images based solely on text, not using any uploaded reference images.\n\nWhen to use `image_gen_oai`:\n- To create entirely new images from detailed text descriptions that do NOT reference any image files.\n\nWhen NOT to use `image_gen_oai`:\n- If the user has uploaded any images and requests modifications, enhancements, or remixing based on those uploads → use `image_edit_oai` instead.\n\nGenerated image IDs will be returned in the response, so you can refer to them in future requests made to `image_edit_oai`.",
        "parameters": {
          "type": "object",
          "properties": {
            "prompt": {
              "type": "string",
              "maxLength": 32000,
              "description": "Describe the image you want in detail. \n      Be highly specific—break your idea into layers: \n      (1) main concept and subject,\n      (2) composition and position,\n      (3) lighting and mood,\n      (4) style, medium, or camera details,\n      (5) important features (age, expression, clothing, etc.),\n      (6) background.\n      Use positive, descriptive language and specify what should be included, not what to avoid. \n      List number and characteristics of people/objects, and mention style/technical requirements (e.g., \"DSLR photo, 85mm lens, golden hour\").\n      Do not reference any uploaded images—use for new image creation from text only."
            },
            "background": {
              "type": "string",
              "enum": [
                "transparent",
                "opaque",
                "auto"
              ],
              "description": "Sets transparency for the background. Must be one of transparent, opaque or auto (default). When transparent, the output format should be png or webp."
            },
            "quality": {
              "type": "string",
              "enum": [
                "auto",
                "high",
                "medium",
                "low"
              ],
              "description": "The quality of the image. One of auto (default), high, medium, or low."
            },
            "size": {
              "type": "string",
              "enum": [
                "auto",
                "2560x1440",
                "1440x2560",
                "3072x2048",
                "2048x3072",
                "2576x1104",
                "1104x2576",
                "2048x2048",
                "3840x2160",
                "2160x3840",
                "3264x2176",
                "2176x3264",
                "3808x1632",
                "1632x3808",
                "2880x2880",
                "1376x768",
                "768x1376",
                "1536x1024",
                "1344x576",
                "576x1344",
                "1024x1536",
                "1024x1024"
              ],
              "description": "The size of the generated image. One of 1024x1024, 1536x1024 (landscape), 1024x1536 (portrait), or auto (default)."
            }
          },
          "required": [
            "prompt"
          ]
        }
      }
    },
    {
      "type": "function",
      "function": {
        "name": "image_edit_oai",
        "description": "Generates high-quality, original images based on text and one or more uploaded/referenced images.\n\nWhen to use `image_edit_oai`:\n- The user wants to modify, extend, or remix one **or more** uploaded images, either:\n- Previously generated, or in the current request (both to be included in the `image_ids` array).\n- Always when the user refers to uploaded images for editing, enhancement, remixing, style transfer, or combining elements.\n- Any current or existing images are to be used as visual guides.\n- If there are any files in the current request, they are more likely than not expected as references for image edit requests.\n\nWhen NOT to use `image_edit_oai`:\n- Brand-new generations that do not rely on an existing image → use `image_gen_oai` instead.\n\nBoth generated and referenced image IDs will be returned in the response, so you can refer to them in future requests made to `image_edit_oai`.",
        "parameters": {
          "type": "object",
          "properties": {
            "image_ids": {
              "type": "array",
              "items": {
                "type": "string"
              },
              "minItems": 1,
              "description": "IDs (image ID strings) of previously generated or uploaded images that should guide the edit.\n\nGuidelines:\n- If the user's request depends on any prior image(s), copy their image IDs into the `image_ids` array (in the same order the user refers to them).  \n- Never invent or hallucinate IDs; only use IDs that are still visible in the conversation context.\n- If no earlier image is relevant, omit the field entirely."
            },
            "prompt": {
              "type": "string",
              "maxLength": 32000,
              "description": "Describe the changes, enhancements, or new ideas to apply to the uploaded image(s).\n      Be highly specific—break your request into layers: \n      (1) main concept or transformation,\n      (2) specific edits/replacements or composition guidance,\n      (3) desired style, mood, or technique,\n      (4) features/items to keep, change, or add (such as objects, people, clothing, lighting, etc.).\n      Use positive, descriptive language and clarify what should be included or changed, not what to avoid.\n      Always base this prompt on the most recently uploaded reference images."
            },
            "quality": {
              "type": "string",
              "enum": [
                "auto",
                "high",
                "medium",
                "low"
              ],
              "description": "The quality of the image. One of auto (default), high, medium, or low. High/medium/low only supported for gpt-image-1."
            },
            "size": {
              "type": "string",
              "enum": [
                "auto",
                "2560x1440",
                "1440x2560",
                "3072x2048",
                "2048x3072",
                "2576x1104",
                "1104x2576",
                "2048x2048",
                "3840x2160",
                "2160x3840",
                "3264x2176",
                "2176x3264",
                "3808x1632",
                "1632x3808",
                "2880x2880",
                "1376x768",
                "768x1376",
                "1536x1024",
                "1344x576",
                "576x1344",
                "1024x1536",
                "1024x1024"
              ],
              "description": "The size of the generated images. For gpt-image-1: auto (default), 1024x1024, 1536x1024, 1024x1536. For dall-e-2: 256x256, 512x512, 1024x1024."
            }
          },
          "required": [
            "image_ids",
            "prompt"
          ]
        }
      }
    }
  ],
  "store": true,
  "reasoning_effort": "high",
  "max_tokens": 64000,
  "messages": [
    {
      "role": "system",
      "content": "你是xxx"
    },
    {
      "role": "user",
      "content": "## Imager 任务 · xxxx ## 生成参数\n- size: 2576x1104"
    },
    {
      "role": "assistant",
      "content": "",
      "tool_calls": [
        {
          "id": "call_8adaea105db647dfbf932053",
          "type": "function",
          "function": {
            "name": "query_structured",
            "arguments": "{\"query\":\"GPT-Image-2\"}"
          }
        },
        {
          "id": "call_32c520f4f2c34bb1b7783fdd",
          "type": "function",
          "function": {
            "name": "query_structured",
            "arguments": "{\"query\":\"prompt\"}"
          }
        }
      ]
    },
    {
      "role": "tool",
      "content": "问题：\nGPT-Image-2",
      "name": "query_structured",
      "tool_call_id": "call_8adaea105db647dfbf932053"
    },
    {
      "role": "tool",
      "content": "问题：\nenergy",
      "name": "query_structured",
      "tool_call_id": "call_32c520f4f2c34bb1b7783fdd"
    }
  ]
}

2026 年 5月 12 日 22:35 更新

更新：之前能正常发的请求，现在也发出去也是 The reasoning_content in the thinking mode must be passed back to the API. 了

2026 年 5月 12 日 22:46 更新

测试了回传思考内容，是可以回复了，甚至空值也可以回复。

看来是在改什么内容了，说不定明天 支持视觉的 v3 Pro 就搬上来了

9 个帖子 - 5 位参与者

阅读完整话题