> ## Documentation Index > Fetch the complete documentation index at: https://docs.aiohub.org/llms.txt > Use this file to discover all available pages before exploring further. # Create video ยท Ark / Seedance > Submit a Seedance video task using Ark / Seedance, with examples, input limits and full parameters. Use this format for a new integration or for reference audio, reference video and video editing. See [Video models and pricing](/en/api/video-models). ## Setup Use `https://api.aiohub.org`. Every request carries `Authorization: Bearer `. Replace the placeholder with your full `sk-` API key and choose a key group that can access the model. ## Example request Start with a text-only request so no image setup is needed. ```bash theme={"system"} curl https://api.aiohub.org/v1/contents/generations/tasks \ -H "Authorization: Bearer " \ -H "Content-Type: application/json" \ -d '{ "model": "doubao-seedance-2-5-260628", "content": [{"type": "text", "text": "A blue ceramic cup on a wooden table in daylight."}], "resolution": "720p", "ratio": "16:9", "duration": 5 }' ``` Creation returns an `id`. Save it for polling; the video is not ready yet. ```json theme={"system"} {"id": "task_example"} ``` After saving the task identifier, [poll the task](/en/api/video-ark-get). ## Image, audio and video inputs An Ark / Seedance task request carries a `content[]` array of a text prompt and reference media. Each media item names its `role`: | Media | `role` values | | - | - | | `image_url` | `first_frame` (default when omitted), `last_frame`, `reference_image` | | `video_url` | `reference_video` | | `audio_url` | `reference_audio` | Media `url` values are HTTPS URLs. Images and audio also accept base64 data URLs such as `data:image/png;base64,...`. Each item may be up to 100 MB. | Input | 2.5 | 2.0, 2.0 fast, 2.0 mini | | - | - | - | | Text only | yes | yes | | One `first_frame` image | yes, any ratio | yes, any ratio | | `first_frame` + `last_frame` | yes, `ratio: adaptive` | yes, any ratio | | `reference_image` images only | 1 to 30 | 1 to 9 | | Reference images with videos or audio | up to 30 images, 10 videos, 10 audios | up to 9 images, 3 videos, 3 audios | | `reference_audio` only | 1 to 10 | | | Video edit (`omni_reference_task_type: "edit"`) | one `reference_video`, `ratio: adaptive`, `duration: -1` | | Images sent next to a video or audio need `role: reference_image`. Reference clips are 2 to 30 s each and 30 s in total on 2.5, and 2 to 15 s each and 15 s in total on the 2.0 series. An edit source video is 4 to 30 s and the output follows its length. ## Status callbacks Set `callback_url` at creation to receive the task object by POST on status changes. Requests carry `User-Agent: AIOHub-LibTV-Callback/1`. Respond with 2xx within 5 seconds; failed deliveries are retried up to three times. If creation times out without a task identifier, check the [usage log](https://api.aiohub.org/console/log) before resubmitting to avoid duplicate generation and charges. Correct parameters after a 400 and check the API key after a 401. See [Authentication](/en/api/authentication) for access and quota requirements. ## OpenAPI ````yaml openapi/relay.en.json POST /v1/contents/generations/tasks openapi: 3.0.1 info: title: AIOHub API description: AIOHub public API version: 1.0.0 servers: - url: https://api.aiohub.org security: - BearerAuth: [] tags: - name: List models - name: OpenAI Chat - name: OpenAI Responses - name: Image generation - name: OpenAI Images - name: Video generation - name: Claude Messages - name: Gemini API - name: OpenAI Embeddings - name: Text completions - name: OpenAI Audio - name: Realtime API paths: /v1/contents/generations/tasks: post: tags: - Video generation summary: Create a video task description: > Submit a Seedance video generation task and get back its ID. The task runs asynchronously; read it by ID to poll the status and download the result. Models, resolutions, durations, input combinations and pricing are described in [Video models and pricing](/en/api/video-models). Validation failures return 400 and are not charged. operationId: createSeedanceVideoTask parameters: [] requestBody: content: application/json: schema: $ref: '#/components/schemas/SeedanceCreateTaskRequest' examples: text: summary: Text to video value: model: doubao-seedance-2-5-260628 content: - type: text text: >- A blue ceramic cup on a wooden table in daylight, slow camera push-in. resolution: 720p ratio: '16:9' duration: 5 first_frame: summary: First-frame image to video value: model: doubao-seedance-2-5-260628 content: - type: text text: >- A blue ceramic cup on a wooden table in daylight, slow camera push-in. - type: image_url image_url: url: https://example.com/first.png role: first_frame resolution: 720p ratio: adaptive duration: 5 generate_audio: true reference: summary: Reference image and audio value: model: doubao-seedance-2-5-260628 content: - type: text text: >- A blue ceramic cup on a wooden table in daylight, slow camera push-in. - type: image_url image_url: url: https://example.com/style.png role: reference_image - type: audio_url audio_url: url: https://example.com/voice.wav role: reference_audio resolution: 1080p duration: 8 responses: '200': description: Task created headers: {} content: application/json: schema: type: object properties: id: type: string description: Task ID required: - id example: id: task_ADQ5XK6AKWg36WePoJ3sZRyDvWl1ogGZ '400': description: >- A field or value is outside the supported range. `code` is one of `unsupported_field`, `invalid_request`, `invalid_duration`, `unsupported_resolution`, `unsupported_ratio`, `invalid_media_mode`, `unsupported_media_carrier` or `invalid_media`. content: application/json: schema: $ref: '#/components/schemas/SeedanceTaskError' headers: {} '401': description: Missing or invalid API key content: application/json: schema: $ref: '#/components/schemas/ErrorResponse' headers: {} '403': description: Insufficient balance content: application/json: schema: $ref: '#/components/schemas/ErrorResponse' headers: {} '429': description: Upstream capacity is saturated, retry later content: application/json: schema: $ref: '#/components/schemas/ErrorResponse' headers: {} '503': description: >- The API key's group has no channel for the model; check the key group content: application/json: schema: $ref: '#/components/schemas/ErrorResponse' headers: {} deprecated: false security: - BearerAuth: [] components: schemas: SeedanceCreateTaskRequest: type: object required: - model - content properties: model: type: string example: doubao-seedance-2-5-260628 description: >- Video model name. See [Video generation](/en/api/video#models) for the models and their limits. content: type: array items: $ref: '#/components/schemas/SeedanceContentItem' description: >- Prompt and reference media. Text-to-video needs one `text` item only. resolution: type: string enum: - 480p - 720p - 1080p - 4k default: 720p description: Output resolution. The allowed values depend on the model. ratio: type: string enum: - adaptive - '16:9' - '4:3' - '1:1' - '3:4' - '9:16' - '21:9' default: adaptive description: >- Aspect ratio. `adaptive` lets the model follow the prompt or the input image. duration: type: integer default: 5 description: >- Duration in whole seconds. The allowed range depends on the model. Video edit tasks use `-1`. generate_audio: type: boolean default: true description: Whether to generate audio. omni_reference_task_type: type: string enum: - auto - reference - edit default: auto description: >- How reference media is used. `edit` edits a single `reference_video`. return_last_frame: type: boolean default: false description: >- When `true`, the task result includes `content.last_frame_url`, a JPEG of the final frame. tools: type: array items: type: object properties: type: type: string enum: - web_search description: >- Set to `[{"type": "web_search"}]` to let the model search the web while generating. priority: type: integer minimum: 0 maximum: 9 default: 0 description: >- Queue priority. Higher values dequeue earlier; running tasks are not preempted. callback_url: type: string format: uri description: >- Public HTTPS URL. AIOHub POSTs the task object to it on each status change; reply with 2xx within 5 seconds or the delivery is retried up to three times. safety_identifier: type: string maxLength: 64 description: >- Stable end-user identifier, up to 64 characters, echoed in task reads. SeedanceTaskError: type: object description: Error response of the video task endpoints. properties: code: type: string description: Error code example: invalid_duration message: type: string description: Error message example: duration must be between 4 and 30 seconds for this model ErrorResponse: type: object properties: error: type: object properties: message: type: string description: Error message type: type: string description: Error type param: type: string description: Related parameters nullable: true code: type: string description: Error code nullable: true SeedanceContentItem: type: object description: 'One `content[]` item: a text prompt or a reference media item.' required: - type properties: type: type: string enum: - text - image_url - video_url - audio_url description: Item type. text: type: string description: >- Prompt text, required when `type` is `text`. Several text items are joined with newlines, up to 20000 characters in total. image_url: type: object properties: url: type: string description: >- HTTPS URL, or a base64 data URL such as `data:image/png;base64,...`. Each media item may be up to 100 MB. description: Required when `type` is `image_url`. video_url: type: object properties: url: type: string description: HTTPS URL of the video. Each media item may be up to 100 MB. description: Required when `type` is `video_url`. audio_url: type: object properties: url: type: string description: >- HTTPS URL, or a base64 data URL such as `data:audio/wav;base64,...`. Each media item may be up to 100 MB. description: Required when `type` is `audio_url`. role: type: string enum: - first_frame - last_frame - reference_image - reference_video - reference_audio description: >- How the media is used. Images default to `first_frame`; images sent next to a video or audio need `reference_image`. Videos are always `reference_video` and audio is always `reference_audio`. securitySchemes: BearerAuth: type: http scheme: bearer description: API key (sk-xxx) ```` This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.