> For the complete documentation index, see [llms.txt](https://anyint.gitbook.io/docs/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://anyint.gitbook.io/docs/api-reference/media-apis.md). # media apis AnyInt currently publishes media-generation routes through DashScope. These are not chat-completion endpoints. They have their own request bodies, and video generation is task-based. ## Published routes | Route | Purpose | | ------------------------------------------------------------------- | ---------------------------------- | | `POST /v1/dashscope/services/aigc/multimodal-generation/generation` | Prompt-driven image generation | | `POST /v1/dashscope/services/aigc/video-generation/video-synthesis` | Prompt, image, and audio to video | | `GET /v1/dashscope/tasks/{task_id}` | Query video task status and result | ## Authentication ```http Authorization: Bearer ``` ## DashScope image generation ### Core request shape The published schema expects: * `model` * `input.messages` * `parameters.size` * optional image controls such as `negative_prompt`, `prompt_extend`, and `watermark` ### cURL example ```bash curl https://gateway.api.anyint.ai/v1/dashscope/services/aigc/multimodal-generation/generation \ -H "Authorization: Bearer $ANYINT_API_KEY" \ -H "Content-Type: application/json" \ -d '{ "model": "qwen-image-plus", "input": { "messages": [ { "role": "user", "content": [ { "text": "A cinematic product poster for an AI music launch, neon highlights, premium studio lighting." } ] } ] }, "parameters": { "negative_prompt": "", "prompt_extend": true, "watermark": false, "size": "1328*1328" } }' ``` Useful response fields: * `output.choices[0].message.content[].image` * `usage.width` * `usage.height` * `usage.image_count` ## DashScope video generation ### Core request shape The current schema expects: * `model` * `input.prompt` * `input.img_url` * `input.audio_url` * `parameters.resolution` * `parameters.duration` * `parameters.audio` * `parameters.shot_type` ### cURL example ```bash curl https://gateway.api.anyint.ai/v1/dashscope/services/aigc/video-generation/video-synthesis \ -H "Authorization: Bearer $ANYINT_API_KEY" \ -H "Content-Type: application/json" \ -d '{ "model": "wan2.5-i2v-preview", "input": { "prompt": "A graffiti rapper comes to life under a railway bridge at night and performs to the beat.", "img_url": "https://example.com/source-image.png", "audio_url": "https://example.com/source-audio.mp3" }, "parameters": { "resolution": "720P", "prompt_extend": true, "duration": 10, "audio": true, "shot_type": "multi" } }' ``` ### Response pattern This is a task-based API. The response returns: * `output.task_id` * `output.task_status` * `metadata.provider` ## Video task query The current catalog shows a task query path in the form: `GET /v1/dashscope/tasks/{task_id}` Apifox currently includes one sample task ID directly in the example path. Treat that as a path template rather than a fixed ID. ### Query example ```bash curl https://gateway.api.anyint.ai/v1/dashscope/tasks/your-task-id \ -H "Authorization: Bearer $ANYINT_API_KEY" ``` ### Query response fields | Field | Meaning | | ---------------------- | ----------------------------------------- | | `output.task_status` | Current task state | | `output.video_url` | Generated video URL after success | | `output.actual_prompt` | Prompt used after provider-side expansion | | `usage.duration` | Generated duration | | `usage.video_count` | Number of outputs | In the published examples, video tasks move from `PENDING` to `SUCCEEDED`. ### HappyHorse video generation HappyHorse is available through the DashScope-compatible video synthesis API. Create a video task first, then poll the task endpoint until `output.task_status` becomes `SUCCEEDED`. Use `happyhorse-1.0-t2v` as the model. The request can include a prompt, a source image URL, and a source audio URL. The initial response returns `request_id` for tracing and `output.task_id` for result polling. #### Create a video task ```bash curl --location --request POST 'https://gateway.api.anyint.ai/dashscope/v1/services/aigc/video-generation/video-synthesis' \ --header "Authorization: Bearer $ANYINT_API_KEY" \ --header 'Content-Type: application/json' \ --data-raw '{ "model": "happyhorse-1.0-t2v", "input": { "prompt": "An urban fantasy scene in cinematic style. A graffiti-painted teenage rapper comes alive from a concrete wall under a railway bridge at night, performs a fast English rap, and strikes an energetic classic rapper pose. The only audio is the rap, with no extra dialogue or noise.", "img_url": "https://help-static-aliyun-doc.aliyuncs.com/file-manage-files/zh-CN/20250925/wpimhv/rap.png", "audio_url": "https://help-static-aliyun-doc.aliyuncs.com/file-manage-files/zh-CN/20250925/ozwpvi/rap.mp3" }, "parameters": { "resolution": "720P", "prompt_extend": true, "duration": 10, "audio": true, "shot_type": "multi" } }' Response example: { "request_id": "570dd72c-52ae-9151-b819-411edf66069e", "output": { "task_id": "c25723f0-8f2b-401a-b843-7ae05ae4e5e3", "task_status": "PENDING" } } Query the generated video Use the output.task_id from the previous response: curl --location --request GET "https://gateway.api.anyint.ai/dashscope/v1/tasks/$TASK_ID" \ --header "Authorization: Bearer $ANYINT_API_KEY" Response example: { "request_id": "1e1fb115-c22a-9769-af2c-056451a8f692", "output": { "task_id": "a54b0585-aa49-45be-8786-fc46414634da", "task_status": "SUCCEEDED", "submit_time": "2026-04-27 17:01:39.657", "scheduled_time": "2026-04-27 17:01:39.689", "end_time": "2026-04-27 17:07:10.758", "orig_prompt": "An urban fantasy scene in cinematic style...", "video_url": "https://example.com/generated-video.mp4" }, "usage": { "duration": 10, "input_video_duration": 0, "output_video_duration": 10, "video_count": 1, "SR": 720, "ratio": "16:9" } } ``` Copy output.video\_url into a browser to preview or download the generated video. Video URLs are temporary, so save the file if you need to keep it. ## When to use media APIs vs AI Music Use media APIs for: * image generation * prompt-to-video * image-plus-audio-to-video Use [AI Music](/docs/ai-music/overview.md) for: * song generation * cover workflows * personas * lyrics * stems * music video generation tied to Suno tasks ## Common mistakes * Treating the example task ID in Apifox as a fixed route * Assuming video generation returns the final MP4 in the initial response * Reusing Gemini image-generation payloads against DashScope routes