doubao-seedance-2-0
doubao-seedance-2.0 Video Generation
- Async processing mode, returns task ID for subsequent queries
- Supports text-to-video, image-to-video (first frame/last frame)
- Supports reference video, reference audio, audio-enabled video
- Supports landscape, portrait, square, ultra-wide, and adaptive aspect ratios
POST
Documentation Index
Fetch the complete documentation index at: https://gccai.heqingsong.uk/llms.txt
Use this file to discover all available pages before exploring further.
Authentication
All API endpoints require Bearer Token authenticationGet your API Key:Visit the API Key Management Page to get your API KeyAdd it to the request header:
Request Parameters
Video generation model nameSupported models:
doubao-seedance-2.0- Standard version, supports text-to-video, image-to-video, first/last frame video, reference video, reference audio, and audio-enabled videodoubao-seedance-2.0-fast- Fast version, same features as the standard version with faster generation speeddoubao-seedance-2.0-face- Supports real person uploads, same features as the standard versiondoubao-seedance-2.0-fast-face- Supports real person uploads, same features as the fast version
Video content descriptionRequired for text-to-video; optional for image-to-video or video-reference-to-videoIt is recommended to clearly specify the subject, action, camera movement, and style for better generation resultsExample:
"A kitten yawning at the camera"Video duration (seconds)Supported range:
4 to 15 secondsDefault: 5Video aspect ratioOptions:
16:9- Landscape9:16- Portrait1:1- Square4:3- Traditional ratio3:4- Vertical traditional ratio21:9- Ultra-wideadaptive- Adaptive (automatically matches the input image/video)
16:9Video resolutionOptions:
480p- Standard definition720p- High definition1080p- Full HD (only supported bydoubao-seedance-2.0-faceanddoubao-seedance-2.0)
480pRandom seed for controlling the randomness of generated content
- With the same request, different seed values will produce different results
- With the same request, the same seed value will produce similar results, but exact consistency is not guaranteed
Whether to generate audio (audio-enabled video)When set to
true, the video will include AI-generated accompanying audioDefault: falseWhether to return the last frame imageWhen set to
true, the task result will additionally return the URL of the video’s last frame image, which can be used for continuous video generationDefault: falseTool list for enhanced capabilities such as web searchExample:
[{"type": "web_search"}]Image URL array for image-to-videoSupports two formats:
- Regular image URL:
https://example.com/cat.jpg - Asset URL (approved asset):
asset://asset_a
["https://example.com/cat.jpg"] or ["asset://asset_a"]Asset URL is only supported by
doubao-seedance-2.0 and doubao-seedance-2.0-fast models. Other models do not support it.Image array with roles, supports specifying first frame/last frameExample:Asset URL format:
When the
url field uses an Asset URL, only doubao-seedance-2.0 and doubao-seedance-2.0-fast models are supported. Other models do not support it.Reference video URL arraySupports two formats:
- Regular video URL:
https://example.com/reference.mp4 - Asset URL (approved asset):
asset://asset_a
["https://example.com/reference.mp4"] or ["asset://asset_a"]Asset URL is only supported by
doubao-seedance-2.0 and doubao-seedance-2.0-fast models. Other models do not support it.Reference audio URL arraySupports two formats:
- Regular audio URL:
https://example.com/speech.wav - Asset URL (approved asset):
asset://asset_a
["https://example.com/speech.wav"] or ["asset://asset_a"]Asset URL is only supported by
doubao-seedance-2.0 and doubao-seedance-2.0-fast models. Other models do not support it.Response
Response status code, 200 on success
Response data array
Use Cases
Case 1: Text-to-Video
Case 2: Image-to-Video (First Frame)
Case 3: First/Last Frame Video
Case 4: Video-Reference-to-Video
Case 5: Reference Video + Reference Audio
Case 6: Audio-Enabled Video
Case 7: Continuous Video Generation (Return Last Frame)
Case 8: Fast Version Generation
Case 9: Reference Images + Reference Video + Reference Audio (Multi-Modal Video)
Combine reference images, reference video, and reference audio to generate an immersive first-person perspective advertisement video. Ideal for product promotions, brand ads, and other scenarios requiring multi-source material fusion.Case 10: Image-to-Video with Asset URL
Approved virtual avatar assets can be passed directly as reference images without re-uploading or re-reviewing.Case 11: Specify Reference Portrait with Asset URL (image_with_roles)
Case 12: Fast Version + Asset URL Image-to-Video
Case 13: Asset URL Image + Reference Video (Motion Transfer)
Combine an approved portrait asset with a reference video to drive the character to perform specified movements.Query Task ResultsVideo generation is an async task that returns a
task_id upon submission. Use the Get Task Status endpoint to query generation progress and results.Differences from 1.5 Pro Version
| Feature | 1.5 Pro | 2.0 / 2.0 fast |
|---|---|---|
| Resolution | 480p/720p/1080p | 480p/720p |
| Duration range | 4-12s | 5-15s |
| Default duration | 5s | 5s |
| Aspect ratio parameter | aspect_ratio | size (new adaptive option) |
| Audio generation | audio parameter | generate_audio parameter |
| Reference video | Not supported | Supported via video_urls |
| Reference audio | Not supported | Supported via audio_urls |
| Image-to-video | image_urls / image_with_roles | image_urls / image_with_roles |
| Audio-enabled video | Not supported | Supported via generate_audio |
| Continuous video | Not supported | Supported via return_last_frame |
| Fast version | Not supported | Supported via doubao-seedance-2.0-fast |