图片理解
image-understanding 通过 FUN Token 豆包多模态模型理解一至八张平台图片，异步返回文字或结构化 JSON。
能力信息
| 字段 | 值 |
|---|---|
| 能力 Key | image-understanding |
| 能力名称 | 图片理解 |
| 能力类型 | model_infer |
| 分类 | 多媒体 |
| 开放状态 | 可通过统一开放能力执行入口调用 |
调用方式
POST /api/open/v1/organizations/{orgId}/capabilities/image-understanding:executeAuthorization: Bearer <sk-organization-application-api-key>Content-Type: application/json请求示例
{ "applicationId": "bbbbbbbb-bbbb-bbbb-bbbb-bbbbbbbbbbbb", "input": { "idempotencyKey": "idempotencyKey", "prompt": "prompt", "imageAssetIds": [ "imageAssetIds" ], "responseFormat": "json_object", "maxTokens": 2048 }}响应示例
{ "capabilityKey": "image-understanding", "message": "SUCCEEDED", "code": 200, "data": { "taskId": "taskId", "status": "PENDING", "model": "model", "pollUrl": "pollUrl", "content": "content", "structured": {}, "usage": {}, "finishReason": "finishReason", "error": {}, "createdAt": "createdAt", "updatedAt": "updatedAt" }}本页由实时开放能力目录生成。后台调整能力配置后，文档目录刷新即可反映最新配置。