Skip to main content

​
模型对比

模型分辨率支持文生视频图生视频-首帧图生视频-首尾帧有声视频
doubao-seedance-1-0-pro-250528480p/720p/1080p✅✅✅❌
doubao-seedance-1-0-pro-fast-251015480p/720p/1080p✅✅❌❌
doubao-seedance-1-5-pro-251215480p/720p/1080p✅✅✅✅

​
文生视频

通过文本描述生成视频。
配置说明：
  • 将 api.xxx.com 替换为您的实际 API 域名
  • 将 sk-xxx 替换为您的 API 密钥（可在控制台获取）
curl -X POST https://api.xxx.com/v1/video/generations \
   -H "Content-Type: application/json" \
   -H "Authorization: Bearer sk-xxx" \
   -d '{
     "model": "doubao-seedance-1-0-pro-fast-251015",
     "prompt": "多个镜头。一名侦探进入一间光线昏暗的房间。他检查桌上的线索,手里拿起桌上的某个物品。镜头转向他正在思索。",
     "resolution": "480p",
     "ratio": "16:9",
     "duration": 10,
     "fps": 24,
     "watermark": true,
     "seed": 11,
     "camerafixed": false
   }'

{
  "task_id": "cgt-20251209112555-fdqtx"
}
视频生成为异步任务，您需要使用返回的 task_id 来查询生成结果。

​
参数说明

​
基础参数（必填）

参数名类型说明可选值
modelstring模型名称doubao-seedance-1-0-pro-fast-251015 / doubao-seedance-1-0-pro-250528 / doubao-seedance-1-5-pro-251215
promptstring文本提示词中英文描述，建议≤500字
resolutionstring视频分辨率480p / 720p / 1080p
默认：1080p
ratiostring视频宽高比16:9 / 4:3 / 1:1 / 3:4 / 9:16 / 21:9 / adaptive
durationinteger视频时长（秒）2~12秒

​
可选参数

参数名类型默认值说明适用场景
imagesarray-图片URL数组
• 单张图片 = 首帧
• 两张图片 = 首尾帧		首尾帧支持：doubao-seedance-1-0-pro-250528、doubao-seedance-1-5-pro-251215
fpsinteger-视频帧率24
seedinteger-1随机种子
• -1 表示随机
• 取值范围：[-1, 2³²-1]		全部模型
watermarkbooleanfalse是否添加水印true / false
camerafixedbooleanfalse相机固定
固定镜头，减少运镜幅度		true / false
generate_audiobooleanfalse是否生成音频
生成带声音的视频		仅 doubao-seedance-1-5-pro-251215 支持
draftbooleanfalse样片生成模式
快速生成低分辨率预览		仅 doubao-seedance-1-5-pro-251215 支持
仅支持 480p
draft_task_idstring-样片任务ID
基于样片生成高质量成片		使用样片生成成片时必填
service_tierstring-服务层级
• flex = 离线推理模式		全部模型
return_last_framebooleanfalse返回视频尾帧图像
(暂不支持)		true / false

​
图生视频-首帧

基于首帧图片生成视频。 
curl -X POST https://api.xxx.com/v1/video/generations \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer sk-xxx" \
  -d '{
    "model": "doubao-seedance-1-0-pro-fast-251015",
    "prompt": "女孩抱着狐狸,女孩睁开眼,温柔地看向镜头",
    "images": ["https://ark-project.tos-cn-beijing.volces.com/doc_image/i2v_foxrgirl.png"],
    "resolution": "480p",
    "ratio": "adaptive",
    "duration": 10,
    "fps": 24,
    "watermark": true,
    "seed": 11,
    "camerafixed": false
  }'

{
  "task_id": "cgt-20251209112555-fdqtx"
}

​
图生视频-首尾帧

基于首尾两帧图片生成视频（仅 doubao-seedance-1-0-pro-250528 支持）。 
curl -X POST https://api.xxx.com/v1/video/generations \
   -H "Content-Type: application/json" \
   -H "Authorization: Bearer sk-xxx" \
  -d '{
    "model": "doubao-seedance-1-0-pro-250528",
    "prompt": "360度环绕运镜",
    "images": [
        "https://ark-project.tos-cn-beijing.volces.com/doc_image/seepro_first_frame.jpeg",
        "https://ark-project.tos-cn-beijing.volces.com/doc_image/seepro_last_frame.jpeg"
    ],
    "resolution": "480p",
    "ratio": "adaptive",
    "duration": 10,
    "fps": 24,
    "watermark": true,
    "seed": 11,
    "camerafixed": false
}'

{
  "task_id": "cgt-20251209112555-fdqtx"
}

​
查询视频任务

使用任务ID查询视频生成结果。 
curl https://api.xxx.com/v1/video/generations/{task_id} \
   -H "Content-Type: application/json" \
   -H "Authorization: Bearer sk-xxx"

{
    "code": "success",
    "message": "",
    "data": {
        "task_id": "cgt-20251209173715-6mn48",
        "action": "generate",
        "status": "SUCCESS",
        "fail_reason": "",
        "submit_time": 1765273035,
        "start_time": 1765273037,
        "finish_time": 1765273082,
        "progress": "100%",
        "data": {
            "id": "cgt-20251209173715-6mn48",
            "seed": 11,
            "model": "doubao-seedance-1-0-pro-fast-251015",
            "ratio": "16:9",
            "usage": {
                "total_tokens": 97605,
                "completion_tokens": 97605
            },
            "status": "succeeded",
            "content": {
                "video_url": "https://ark-content-generation-cn-beijing.tos-cn-beijing.volces.com/..."
            },
            "duration": 10,
            "created_at": 1765273035,
            "resolution": "480p",
            "updated_at": 1765273077,
            "service_tier": "default",
            "framespersecond": 24,
            "execution_expires_after": 172800
        }
    }
}

​
任务状态说明

状态说明
PENDING任务排队中
PROCESSING任务处理中
SUCCESS任务成功完成
FAILED任务失败

​
有声视频生成

使用 doubao-seedance-1-5-pro-251215 模型生成带音频的视频。

​
有声视频-图生视频

curl -X POST https://api.xxx.com/v1/video/generations \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer sk-xxx" \
  -d '{
    "model": "doubao-seedance-1-5-pro-251215",
    "prompt": "女孩抱着狐狸，女孩睁开眼，温柔地看向镜头，狐狸友善地抱着，镜头缓缓拉出，女孩的头发被风吹动，可以听到风声",
    "images": ["https://ark-project.tos-cn-beijing.volces.com/doc_image/i2v_foxrgirl.png"],
    "resolution": "480p",
    "ratio": "adaptive",
    "duration": 5,
    "fps": 24,
    "watermark": false,
    "generate_audio": true
  }'

{
  "task_id": "cgt-20251209112555-fdqtx"
}

​
有声视频-首尾帧

curl -X POST https://api.xxx.com/v1/video/generations \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer sk-xxx" \
  -d '{
    "model": "doubao-seedance-1-5-pro-251215",
    "prompt": "图中女孩对着镜头说茄子，360度环绕运镜",
    "images": [
      "https://ark-project.tos-cn-beijing.volces.com/doc_image/seepro_first_frame.jpeg",
      "https://ark-project.tos-cn-beijing.volces.com/doc_image/seepro_last_frame.jpeg"
    ],
    "resolution": "480p",
    "ratio": "adaptive",
    "duration": 5,
    "fps": 24,
    "watermark": false,
    "generate_audio": true
  }'

{
  "task_id": "cgt-20251209112555-fdqtx"
}

​
有声视频-文生视频

curl -X POST https://api.xxx.com/v1/video/generations \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer sk-xxx" \
  -d '{
    "model": "doubao-seedance-1-5-pro-251215",
    "prompt": "无人机以极快速度穿越复杂障碍或自然奇观，带来沉浸式飞行体验",
    "resolution": "480p",
    "ratio": "16:9",
    "duration": 5,
    "fps": 24,
    "watermark": true,
    "camerafixed": false,
    "generate_audio": true
  }'

{
  "task_id": "cgt-20251209112555-fdqtx"
}
使用 generate_audio: true 参数可生成带环境音效的视频，模型会根据提示词和画面内容自动生成匹配的音频轨道。

​
样片生成 (Draft Mode)

使用样片模式快速预览视频效果，确认后再生成高清成片。

​
生成样片

curl -X POST https://api.xxx.com/v1/video/generations \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer sk-xxx" \
  -d '{
    "model": "doubao-seedance-1-5-pro-251215",
    "prompt": "延时摄影，因固定镜头，薄雾升腾汇聚成云海，云海沿山势流动，阳光照射使云海泛金光，云层边缘逐渐清晰",
    "resolution": "480p",
    "ratio": "16:9",
    "duration": 5,
    "fps": 24,
    "watermark": false,
    "camerafixed": false,
    "draft": true,
    "generate_audio": true
  }'

{
  "task_id": "cgt-20260121161956-vbk57"
}
  • 样片模式仅支持 480p 分辨率
  • 生成速度更快，成本更低
  • 适合内容预览和效果确认

​
基于样片生成成片

curl -X POST https://api.xxx.com/v1/video/generations \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer sk-xxx" \
  -d '{
    "model": "doubao-seedance-1-5-pro-251215",
    "prompt": "延时摄影，因固定镜头，薄雾升腾汇聚成云海，云海沿山势流动，阳光照射使云海泛金光，云层边缘逐渐清晰",
    "resolution": "720p",
    "ratio": "16:9",
    "duration": 5,
    "draft_task_id": "cgt-20260121161956-vbk57"
  }'

{
  "task_id": "cgt-20260121162156-abc12"
}
  • 使用样片任务ID生成高清成片
  • 支持 720p / 1080p 分辨率
  • 继承样片的内容和风格，提高生成成功率

​
离线推理 (Flex Tier)

适合批量处理场景的异步推理模式。 
curl -X POST https://api.xxx.com/v1/video/generations \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer sk-xxx" \
  -d '{
    "model": "doubao-seedance-1-5-pro-251215",
    "prompt": "延时摄影，因固定镜头，薄雾升腾汇聚成云海，云海沿山势流动，阳光照射使云海泛金光，云层边缘逐渐清晰",
    "resolution": "720p",
    "ratio": "16:9",
    "duration": 5,
    "fps": 24,
    "watermark": false,
    "camerafixed": false,
    "service_tier": "flex",
    "generate_audio": true
  }'

{
  "task_id": "cgt-batch-20260121162256-xyz89"
}
  • 任务ID格式为 cgt-batch-*
  • 成本更低，适合大批量生成
  • 处理时间可能较长，需要等待几分钟到几小时，不等适合非紧急任务
视频生成任务的执行时间取决于队列长度和视频时长，通常需要几分钟时间。