Seedance 2 Video Generation
4s minimum, from 464.52 credits
ae_seedance_2_video_generation
Create a 4–15 second video from a text prompt with optional image, video, or audio references. Choose from Seedance 2.5, Seedance 2.0, or Seedance 2.0 Fast, select 720p or 1080p resolution, and configure aspect ratio, synchronized audio, watermarking, seed, and optional last-frame output.
4s minimum, from 464.52 credits
Pricing
| Model | Resolution: 720p | Resolution: 1080p |
|---|---|---|
| Seedance 2.5 | 219.71 credits/second | 476.82 credits/second |
| Seedance 2.0 | 144.38 credits/second | 313.34 credits/second |
| Seedance 2.0 Fast | 116.13 credits/second | 252.03 credits/second |
The starting price uses video duration (seconds) 4. Starting price applies to Model: Seedance 2.0 Fast, Resolution: 720p.
Parameters
Optional parameters use the platform defaults when omitted.
| Parameter | Type | Required | Description | Default |
|---|---|---|---|---|
seed | integer | No | Random seed. Use -1 for a random seed. | -1 |
model | enum (doubao-seedance-2-5 | doubao-seedance-2-0 | doubao-seedance-2-0-fast) | Yes | 选择 Seedance 2.5、2.0 或 2.0 Fast。模型区别:doubao-seedance-2-5 的首帧/首尾帧模式必须使用 ratio="adaptive",比例跟随首帧;不要传 16:9、9:16 等固定比例。doubao-seedance-2-0 和 doubao-seedance-2-0-fast 的单首帧模式已实测支持 adaptive 和 16:9;不把 2.5 的强制规则套到这两款模型。未实测的其他比例或首尾帧组合不作保证。首帧/末帧内容不能与 reference_image、reference_video、reference_audio 混用,媒体角色应明确填写。 | — |
ratio | enum (16:9 | 4:3 | 1:1 | 3:4 | 9:16 | 21:9 | adaptive) | No | 输出视频宽高比;默认 adaptive。模型区别:doubao-seedance-2-5 的首帧/首尾帧模式必须使用 ratio="adaptive",比例跟随首帧;不要传 16:9、9:16 等固定比例。doubao-seedance-2-0 和 doubao-seedance-2-0-fast 的单首帧模式已实测支持 adaptive 和 16:9;不把 2.5 的强制规则套到这两款模型。未实测的其他比例或首尾帧组合不作保证。首帧/末帧内容不能与 reference_image、reference_video、reference_audio 混用,媒体角色应明确填写。 | adaptive |
content | array | Yes | 多模态生成输入,至少包含一项非空 text。图片、视频和音频可选;音频不能作为唯一参考模态;不支持 draft_task。模型区别:doubao-seedance-2-5 的首帧/首尾帧模式必须使用 ratio="adaptive",比例跟随首帧;不要传 16:9、9:16 等固定比例。doubao-seedance-2-0 和 doubao-seedance-2-0-fast 的单首帧模式已实测支持 adaptive 和 16:9;不把 2.5 的强制规则套到这两款模型。未实测的其他比例或首尾帧组合不作保证。首帧/末帧内容不能与 reference_image、reference_video、reference_audio 混用,媒体角色应明确填写。 | — |
content[].role | string | No | 图片、视频、音频的媒体角色,例如 reference_image、first_frame、last_frame、reference_video、reference_audio。角色支持范围以所选模型为准。模型区别:doubao-seedance-2-5 的首帧/首尾帧模式必须使用 ratio="adaptive",比例跟随首帧;不要传 16:9、9:16 等固定比例。doubao-seedance-2-0 和 doubao-seedance-2-0-fast 的单首帧模式已实测支持 adaptive 和 16:9;不把 2.5 的强制规则套到这两款模型。未实测的其他比例或首尾帧组合不作保证。首帧/末帧内容不能与 reference_image、reference_video、reference_audio 混用,媒体角色应明确填写。 | — |
content[].text | string | No | Text prompt used when type is text. | — |
content[].type | enum (text | image_url | video_url | audio_url) | Yes | No parameter description. | — |
content[].audio_url | object | No | No parameter description. | — |
content[].audio_url.url | string | Yes | Public URL, data URL, or asset URL for the input audio. | — |
content[].image_url | object | No | No parameter description. | — |
content[].image_url.url | string | Yes | Public URL, data URL, or asset URL for the input image. | — |
content[].video_url | object | No | No parameter description. | — |
content[].video_url.url | string | Yes | Public URL, data URL, or asset URL for the input video. | — |
duration | integer | Yes | Output video duration in seconds. Only integer values from 4 to 15 are supported. | — |
watermark | boolean | No | Whether to add an AI-generated watermark. | false |
resolution | enum (720p | 1080p) | Yes | Output video resolution. Supports 720p and 1080p. | — |
generate_audio | boolean | No | Whether to generate synchronized speech, sound effects, and background audio. | false |
return_last_frame | boolean | No | Whether to return the last frame image. | false |
Invocation example
Copy the request and explore its response example.
POST
/agent-api/v1/tool/executeX-Api-KeySubmit first, then query with the returned task_id until the task completes.
Request coderequest.sh
curl --request POST \
--url 'https://agentearth.ai/agent-api/v1/tool/execute?tool_name=ae_seedance_2_video_generation' \
--header "X-Api-Key: $AGENTEARTH_API_KEY" \
--header 'Content-Type: application/json' \
--data-raw '{
"params": {
"content": [
{
"text": "3D Chinese animation film style, cinematic. Spacious shopping mall atrium, glass ceiling, skylight, brand storefronts. A young man sits on a bench scrolling his phone. He suddenly looks up, eyes wide, jumps to his feet and shouts excitedly. Camera starts with a wide establishing shot of the atrium, then pushes in to a close-up of his excited face. Single continuous shot. No text, no subtitles, no watermark.",
"type": "text"
},
{
"image_url": {
"url": "https://agentearth.ai/fileserver/file/ef03a249-ba2d-4a79-bbac-cbd3f6d55bf2.png"
},
"role": "reference_image",
"type": "image_url"
}
],
"duration": 15,
"model": "doubao-seedance-2-0-fast",
"ratio": "9:16",
"resolution": "720p"
}
}'Response exampleresponse.json
{
"description": "task accepted; poll the result via query_tool with this task_id",
"query_tool": "ae_get_result",
"status": "created",
"task_id": "f9c5539e-74ec-49d6-a482-8d86ddc29f16"
}This example shows part of the response. Copied content is valid JSON.
Guides using this tool
See how this tool fits into real workflows
Reference-Guided Video Generation with Character ConsistencyGenerate AI video from character reference images so the same person keeps their face, outfit, and look across shots — no face-swapping for serial IP content, with multiple shots stitched into one continuous clip.What AI Agents Can Do With Tools: 15 Real Use CasesAn AI agent on its own can only talk. Connect it to tools and it can research, source products, generate images and video, pull live financial data, and more. Here are 15 real things agents do with tools.AI Video Generation Workflow with AI AgentsRun text-to-video, image-to-video, and reference-keyframe video generation through one AI agent: pick the path by your starting material, compare models, and choose the best cut — no separate sign-ups.






