Create a Seedance video
Create a Seedance video task on CometAPI. Use text prompts or reference images, select a duration, and request an exact WxH output size.
WxH value from the table below for the size field.Supported model IDs
Use one of these exact values for themodel field.
Use reference images
For Seedance 2.5 image-to-video requests, upload the reference image through the multipartinput_reference field. The example on this page uses one file.
Seedance 2.0, Seedance 2.0 Fast, and Seedance 2.0 Mini also accept
input_reference. For advanced multi-reference prompts with those 2.0 models,
repeat the same multipart field in upload order.
One Mini production request with a single input_reference completed at the 720x1280 output size.
That check confirmed request completion, but it did not isolate the reference image’s visual effect from the prompt.
Seedance 2.0 is sensitive to how the prompt connects each reference image to the scene. Refer to the uploaded files by order, such as [Image 1], [Image 2], and [Image 3], then assign each image a clear role:
- Use
[Image 1]for the main subject, character, or product. - Use
[Image 2]for a secondary subject, prop, or companion. - Use
[Image 3]for the background, setting, lighting, or style.
Duration by model
If you omitseconds, CometAPI requests a 5-second clip. Send seconds as a
string when you need a specific duration.
Size support by model
Seedance 2.5
The following exact sizes apply to both text-to-video and image-to-video requests withseedance-2-5-260628.
Earlier Seedance models
The table below keeps the exactWxH values in one place. Seedance 2.0 Fast
uses only the 480p and 720p values in the fifth column. The 1080p values
in that column apply to Seedance 1.5 Pro and standard Seedance 2.0. The 4K
values apply only to standard Seedance 2.0. In the Mini column, — means the
resolution is not supported.
WxH value for the target model. Exact WxH support remains model-dependent, so inspect the completed media before relying on an undocumented size.Authorizations
Bearer token authentication. Use your CometAPI key.
Body
Text prompt that describes the video. Required. When using reference images, describe what each uploaded image should control, such as the subject from [Image 1], the secondary object from [Image 2], and the background or style from [Image 3]. Also state the action, camera motion, visual style, and scene.
"A slow cinematic camera push across a coastal landscape at sunrise."
Seedance model ID. Use one of the exact values below. Aliases such as seedance-mini are not accepted.
seedance-2-5, doubao-seedance-2-0, doubao-seedance-2-0-fast, doubao-seedance-2-0-mini, doubao-seedance-1-5-pro, doubao-seedance-1-0-pro "doubao-seedance-2-0-mini"
Video duration in seconds. Seedance 2.5 accepts 4 to 30. Seedance 2.0, Seedance 2.0 Fast, and Seedance 2.0 Mini accept 4 to 15. Seedance 1.5 Pro accepts 4 to 12, and Seedance 1.0 Pro accepts 2 to 10. The default is 5.
2 <= x <= 305
Output size as an exact WxH value. For Seedance 2.5 text-to-video and image-to-video, use 854x480, 752x560, 640x640, 560x752, 480x854, or 992x432 for 480p; use 1280x720, 1112x834, 960x960, 834x1112, 720x1280, or 1470x630 for 720p. For Seedance 2.0 Mini, use 864x496, 752x560, 640x640, 560x752, 496x864, or 992x432 for 480p; use the same documented 720p values. The standard Seedance 2.0 model also accepts its documented 4K values.
^[1-9]\d{2,3}x[1-9]\d{2,3}$"854x480"
"480x854"
"864x496"
"752x560"
"640x640"
"560x752"
"496x864"
"992x432"
"1280x720"
"1112x834"
"960x960"
"834x1112"
"720x1280"
"1470x630"
"1920x1080"
"1080x1920"
"3840x2160"
"3326x2494"
"2880x2880"
"2494x3326"
"2160x3840"
"4398x1886"
Optional reference image uploaded as a multipart file. Seedance 2.5 image-to-video requests use this field; the included example uploads one file. Seedance 2.0, Seedance 2.0 Fast, and Seedance 2.0 Mini also accept this field; for their multi-reference requests, repeat the field and reference the files in upload order as [Image 1], [Image 2], and so on. Seedance 1.0 Pro and 1.5 Pro do not accept it.
Response
Task created. Save the returned id and poll GET /v1/videos/{id}.
Task id. Use it as the path parameter for GET /v1/videos/{id}.
Object type, always video.
Echo of the requested model id.
Initial task status. Newly created tasks are returned as queued.
queued, in_progress, completed, failed, error Completion percentage. 0 at creation.
0 <= x <= 100Task creation time as a Unix timestamp in seconds.
Alias of id returned for compatibility. The value matches id.