Back to Details

seedance2-api

Compare original and translation side by side

🇺🇸

Original

English
🇨🇳

Translation

Chinese

Seedance 2.0 分镜创作与视频生成

Seedance 2.0 Storyboard Creation & Video Generation

从创意到成片的全流程：引导分镜 → 画参考图 → 提交视频 → 获取结果。
End-to-end workflow from idea to final video: Storyboard Guidance → Reference Image Generation → Video Submission → Result Retrieval.

Step 0: 确定执行方式（MCP 或脚本）

Step 0: Determine Execution Mode (MCP or Script)

优先检测 MCP 是否可用：
  1. 检查 
    xskill-ai
    MCP 服务状态（读取 
    mcps/user-xskill-ai/STATUS.md
    ）
  2. 如果 MCP 可用 → 使用 
    submit_task
    / 
    get_task
    等 MCP 工具
  3. 如果 MCP 不可用或报错 → 切换到脚本模式
脚本模式前置条件：
  1. 确认环境变量 
    XSKILL_API_KEY
    已设置（Shell 执行 
    echo $XSKILL_API_KEY | head -c 10
    ）
  2. 如果未设置，提示用户： 
    export XSKILL_API_KEY=sk-your-api-key
获取 API Key: https://www.xskill.ai/#/v2/api-keys
  3. 确认 
    requests
    已安装（
    pip install requests
    ）
脚本路径： 本 Skill 目录下的 
scripts/seedance_api.py
，定位方式：
bash
undefined
Prioritize checking MCP availability:
  1. Check the status of the 
    xskill-ai
    MCP service (read 
    mcps/user-xskill-ai/STATUS.md
    )
  2. If MCP is available → Use MCP tools like 
    submit_task
    / 
    get_task
  3. If MCP is unavailable or errors occur → Switch to Script Mode
Prerequisites for Script Mode:
  1. Confirm that the environment variable 
    XSKILL_API_KEY
    is set (execute 
    echo $XSKILL_API_KEY | head -c 10
    in Shell)
  2. If not set, prompt the user: 
    export XSKILL_API_KEY=sk-your-api-key
Get API Key: https://www.xskill.ai/#/v2/api-keys
  3. Confirm 
    requests
    is installed (
    pip install requests
    )
Script Path: 
scripts/seedance_api.py
in this Skill directory, locate it via:
bash
undefined

通过 Glob 工具搜索

Search via Glob tool

glob: .cursor/skills/seedance2-api/scripts/seedance_api.py

> 后续步骤中，每个 API 调用都同时给出 **MCP 方式** 和 **脚本方式**，根据 Step 0 的判断选择其一执行。
glob: .cursor/skills/seedance2-api/scripts/seedance_api.py

> In subsequent steps, each API call provides both **MCP Method** and **Script Method**, choose one based on the judgment from Step 0.

Step 1: 理解用户想法

Step 1: Understand User's Idea

收集以下信息（缺失的主动询问）：
  • 故事核心：一句话概括要拍什么
  • 时长：4-15 秒
  • 画面比例：16:9 / 9:16 / 1:1 / 21:9 / 4:3 / 3:4
  • 视觉风格：写实/动画/水墨/科幻/赛博朋克等
  • 素材情况：是否有现成图片/视频/音频，还是需要 AI 生成
  • 功能模式：是否需要首尾帧控制（first_last_frames），否则默认全能模式（omni_reference）
Collect the following information (proactively ask if missing):
  • Core Story: One-sentence summary of what the video is about
  • Duration: 4-15 seconds
  • Aspect Ratio: 16:9 / 9:16 / 1:1 / 21:9 / 4:3 / 3:4
  • Visual Style: Realistic/Animated/Ink Wash/Sci-Fi/Cyberpunk, etc.
  • Material Status: Whether there are existing images/videos/audio, or if AI generation is needed
  • Function Mode: Whether first-last frame control (
    first_last_frames
    ) is required; otherwise, default to omni-reference mode (
    omni_reference
    )

Step 2: 深入挖掘（5 个维度）

Step 2: Deep Dive (5 Dimensions)

针对每个维度引导用户补充细节：
  1. 内容 - 主角是谁？做什么？在哪里？
  2. 视觉 - 光影、色调、质感、氛围
  3. 镜头 - 推/拉/摇/移/跟/环绕/升降
  4. 动作 - 主体的具体动作和节奏
  5. 声音 - 配乐风格、音效、对白
Guide the user to supplement details for each dimension:
  1. Content - Who is the protagonist? What are they doing? Where is it set?
  2. Visuals - Lighting, color tone, texture, atmosphere
  3. Camera - Push/Pull/Pan/Truck/Follow/Orbit/Crane
  4. Action - Specific actions and rhythm of the subject
  5. Audio - Soundtrack style, sound effects, dialogue

Step 3: 构建分镜结构

Step 3: Build Storyboard Structure

按时间轴拆分镜头，使用以下公式：
【风格】_____风格，_____秒，_____比例，_____氛围

0-X秒：[镜头运动] + [画面内容] + [动作描述]
X-Y秒：[镜头运动] + [画面内容] + [动作描述]
...

【声音】_____配乐 + _____音效 + _____对白
【参考】@image_file_1 _____，@video_file_1 _____
详细模板和示例见 reference.md。
Split shots by timeline using the following formula:
【Style】_____ style, _____ seconds, _____ aspect ratio, _____ atmosphere

0-X seconds: [Camera Movement] + [Frame Content] + [Action Description]
X-Y seconds: [Camera Movement] + [Frame Content] + [Action Description]
...

【Audio】_____ soundtrack + _____ sound effects + _____ dialogue
【Reference】@image_file_1 _____, @video_file_1 _____
Detailed templates and examples can be found in reference.md.

Step 4: 生成参考图（如需要）

Step 4: Generate Reference Images (If Needed)

若用户没有现成素材，使用 Seedream 4.5 生成角色图、场景图、首帧/尾帧等。
If the user has no existing materials, use Seedream 4.5 to generate character images, scene images, first/last frames, etc.

文生图

Text-to-Image

<details> <summary><b>MCP 方式</b></summary>
调用 
submit_task
工具：
  • model_id: 
    fal-ai/bytedance/seedream/v4.5/text-to-image
  • parameters:
    • prompt: 详细的图片描述（英文效果更佳）
    • image_size: 根据视频比例选择
    • num_images: 需要的数量（1-6）
</details> <details> <summary><b>脚本方式</b></summary>
bash
python .cursor/skills/seedance2-api/scripts/seedance_api.py submit \
  --model "fal-ai/bytedance/seedream/v4.5/text-to-image" \
  --params '{"prompt":"An astronaut in a white spacesuit...","image_size":"landscape_16_9","num_images":1}'
</details>
<details> <summary><b>MCP Method</b></summary>
Call the 
submit_task
tool:
  • model_id: 
    fal-ai/bytedance/seedream/v4.5/text-to-image
  • parameters:
    • prompt: Detailed image description (English works better)
    • image_size: Select based on video aspect ratio
    • num_images: Number of images needed (1-6)
</details> <details> <summary><b>Script Method</b></summary>
bash
python .cursor/skills/seedance2-api/scripts/seedance_api.py submit \
  --model "fal-ai/bytedance/seedream/v4.5/text-to-image" \
  --params '{"prompt":"An astronaut in a white spacesuit...","image_size":"landscape_16_9","num_images":1}'
</details>

图像编辑（基于现有图片修改）

Image Editing (Modify Based on Existing Images)

<details> <summary><b>MCP 方式</b></summary>
调用 
submit_task
工具：
  • model_id: 
    fal-ai/bytedance/seedream/v4.5/edit
  • parameters:
    • prompt: 编辑指令（用 Figure 1/2/3 引用图片）
    • image_urls: 输入图片 URL 数组
    • image_size: 输出尺寸
</details> <details> <summary><b>脚本方式</b></summary>
bash
python .cursor/skills/seedance2-api/scripts/seedance_api.py submit \
  --model "fal-ai/bytedance/seedream/v4.5/edit" \
  --params '{"prompt":"Change the background to a forest","image_urls":["https://..."],"image_size":"landscape_16_9"}'
</details>
<details> <summary><b>MCP Method</b></summary>
Call the 
submit_task
tool:
  • model_id: 
    fal-ai/bytedance/seedream/v4.5/edit
  • parameters:
    • prompt: Editing instructions (refer to images using Figure 1/2/3)
    • image_urls: Array of input image URLs
    • image_size: Output size
</details> <details> <summary><b>Script Method</b></summary>
bash
python .cursor/skills/seedance2-api/scripts/seedance_api.py submit \
  --model "fal-ai/bytedance/seedream/v4.5/edit" \
  --params '{"prompt":"Change the background to a forest","image_urls":["https://..."],"image_size":"landscape_16_9"}'
</details>

轮询图片结果

Poll for Image Results

图片约 1-2 分钟完成。
<details> <summary><b>MCP 方式</b></summary>
调用 
get_task
工具查询状态：
  • 首次 30 秒后查询
  • 之后每 30 秒查询一次
  • 状态为 
    completed
    时提取图片 URL
</details> <details> <summary><b>脚本方式</b></summary>
单次查询：
bash
python .cursor/skills/seedance2-api/scripts/seedance_api.py query \
  --task-id "TASK_ID_HERE"
自动轮询（推荐用于图片，间隔 10s，超时 180s）：
bash
python .cursor/skills/seedance2-api/scripts/seedance_api.py poll \
  --task-id "TASK_ID_HERE" --interval 10 --timeout 180
</details>
Images take approximately 1-2 minutes to complete.
<details> <summary><b>MCP Method</b></summary>
Call the 
get_task
tool to query status:
  • First query after 30 seconds
  • Then query every 30 seconds thereafter
  • Extract image URLs when status is 
    completed
</details> <details> <summary><b>Script Method</b></summary>
Single Query:
bash
python .cursor/skills/seedance2-api/scripts/seedance_api.py query \
  --task-id "TASK_ID_HERE"
Auto Polling (Recommended for Images, 10s interval, 180s timeout):
bash
python .cursor/skills/seedance2-api/scripts/seedance_api.py poll \
  --task-id "TASK_ID_HERE" --interval 10 --timeout 180
</details>

image_size 对照表

image_size Reference Table

视频比例推荐 image_size说明
16:9landscape_16_9横屏
9:16portrait_16_9竖屏
4:3landscape_4_3横屏
3:4portrait_4_3竖屏
1:1square_hd方形
21:9landscape_16_9近似超宽屏
Video Aspect RatioRecommended image_sizeDescription
16:9landscape_16_9Landscape
9:16portrait_16_9Portrait
4:3landscape_4_3Landscape
3:4portrait_4_3Portrait
1:1square_hdSquare
21:9landscape_16_9Approximate ultra-wide screen

Step 5: 生成专业提示词

Step 5: Generate Professional Prompts

将分镜结构和参考图整合为最终提示词：
  • 用 
    @image_file_1
    、
    @image_file_2
    等引用 image_files 数组中的图片
  • 用 
    @video_file_1
    等引用 video_files 数组中的视频
  • 用 
    @audio_file_1
    等引用 audio_files 数组中的音频
  • 也兼容旧版 
    @图片1
    、
    @视频1
    、
    @音频1
    语法
引用语法示例：
@image_file_1 作为角色形象参考，参考 @video_file_1 的运镜方式，配合 @audio_file_1 的配乐
重要： 
image_files
数组中第 N 个 URL 对应 
@image_file_N
，
video_files
和 
audio_files
分别独立编号。
Integrate the storyboard structure and reference images into the final prompt:
  • Use 
    @image_file_1
    , 
    @image_file_2
    , etc. to reference images in the 
    image_files
    array
  • Use 
    @video_file_1
    , etc. to reference videos in the 
    video_files
    array
  • Use 
    @audio_file_1
    , etc. to reference audio in the 
    audio_files
    array
  • Also compatible with the old syntax: 
    @图片1
    , 
    @视频1
    , 
    @音频1
Reference Syntax Example:
@image_file_1 as character reference, refer to the camera movement of @video_file_1, match the soundtrack of @audio_file_1
Important: The Nth URL in the 
image_files
array corresponds to 
@image_file_N
. 
video_files
and 
audio_files
are numbered independently.

Step 6: 提交视频任务

Step 6: Submit Video Task

处理素材 URL：
  • Seedream 生成的图片：已有 URL，直接使用
  • 用户提供的网络图片：直接使用
  • 用户提供的本地图片：先上传获取 URL（见下方上传方式）
Process Material URLs:
  • Images generated by Seedream: URLs already exist, use directly
  • Web images provided by user: Use directly
  • Local images provided by user: Upload first to get URL (see upload methods below)

上传本地图片

Upload Local Images

<details> <summary><b>MCP 方式</b></summary>
调用 
upload_image
工具：image_url 或 image_data
</details> <details> <summary><b>脚本方式</b></summary>
bash
undefined
<details> <summary><b>MCP Method</b></summary>
Call the 
upload_image
tool: image_url or image_data
</details> <details> <summary><b>Script Method</b></summary>
bash
undefined

上传网络图片

Upload web image

python .cursor/skills/seedance2-api/scripts/seedance_api.py upload
--image-url "https://example.com/image.png"
python .cursor/skills/seedance2-api/scripts/seedance_api.py upload
--image-url "https://example.com/image.png"

上传本地图片

Upload local image

python .cursor/skills/seedance2-api/scripts/seedance_api.py upload
--image-path "/path/to/local/image.png"

</details>
python .cursor/skills/seedance2-api/scripts/seedance_api.py upload
--image-path "/path/to/local/image.png"

</details>

提交 Seedance 2.0 任务（全能模式 omni_reference）

Submit Seedance 2.0 Task (Omni-reference Mode 
omni_reference
)

<details> <summary><b>MCP 方式</b></summary>
调用 
submit_task
工具：
  • model_id: 
    st-ai/super-seed2
  • parameters:
    • prompt: Step 5 生成的完整提示词
    • functionMode: 
      omni_reference
      （默认，可省略）
    • image_files: 参考图片 URL 数组（最多 9 张，顺序对应 @image_file_1/2/3...）
    • video_files: 参考视频 URL 数组（最多 3 个，总时长 ≤ 15s）
    • audio_files: 参考音频 URL 数组（最多 3 个）
    • ratio: 画面比例（
      16:9
      / 
      9:16
      / 
      1:1
      / 
      21:9
      / 
      4:3
      / 
      3:4
      ）
    • duration: 时长整数（
      4
      - 
      15
      ）
    • model: 
      seedance_2.0_fast
      （默认快速）或 
      seedance_2.0
      （标准质量）
</details> <details> <summary><b>脚本方式</b></summary>
bash
python .cursor/skills/seedance2-api/scripts/seedance_api.py submit \
  --model "st-ai/super-seed2" \
  --params '{
    "prompt": "电影级写实科幻风格，15秒，16:9...",
    "functionMode": "omni_reference",
    "image_files": ["https://img1.png", "https://img2.png"],
    "ratio": "16:9",
    "duration": 15,
    "model": "seedance_2.0_fast"
  }'
</details>
<details> <summary><b>MCP Method</b></summary>
Call the 
submit_task
tool:
  • model_id: 
    st-ai/super-seed2
  • parameters:
    • prompt: Complete prompt generated in Step 5
    • functionMode: 
      omni_reference
      (default, can be omitted)
    • image_files: Array of reference image URLs (max 9 images, order corresponds to @image_file_1/2/3...)
    • video_files: Array of reference video URLs (max 3, total duration ≤ 15s)
    • audio_files: Array of reference audio URLs (max 3)
    • ratio: Aspect ratio (
      16:9
      / 
      9:16
      / 
      1:1
      / 
      21:9
      / 
      4:3
      / 
      3:4
      )
    • duration: Integer duration (4 - 15)
    • model: 
      seedance_2.0_fast
      (default fast mode) or 
      seedance_2.0
      (standard quality)
</details> <details> <summary><b>Script Method</b></summary>
bash
python .cursor/skills/seedance2-api/scripts/seedance_api.py submit \
  --model "st-ai/super-seed2" \
  --params '{
    "prompt": "Cinematic realistic sci-fi style, 15 seconds, 16:9...",
    "functionMode": "omni_reference",
    "image_files": ["https://img1.png", "https://img2.png"],
    "ratio": "16:9",
    "duration": 15,
    "model": "seedance_2.0_fast"
  }'
</details>

提交 Seedance 2.0 任务（首尾帧模式 first_last_frames）

Submit Seedance 2.0 Task (First-Last Frame Mode 
first_last_frames
)

<details> <summary><b>MCP 方式</b></summary>
调用 
submit_task
工具：
  • model_id: 
    st-ai/super-seed2
  • parameters:
    • prompt: 视频描述提示词
    • functionMode: 
      first_last_frames
    • filePaths: 图片 URL 数组（0 张=文生视频，1 张=首帧图生视频，2 张=首尾帧视频）
    • ratio: 画面比例
    • duration: 时长整数
    • model: 
      seedance_2.0_fast
      或 
      seedance_2.0
</details> <details> <summary><b>脚本方式</b></summary>
bash
python .cursor/skills/seedance2-api/scripts/seedance_api.py submit \
  --model "st-ai/super-seed2" \
  --params '{
    "prompt": "镜头从首帧缓缓过渡到尾帧，画面流畅自然",
    "functionMode": "first_last_frames",
    "filePaths": ["https://first-frame.png", "https://last-frame.png"],
    "ratio": "16:9",
    "duration": 5,
    "model": "seedance_2.0_fast"
  }'
</details>
<details> <summary><b>MCP Method</b></summary>
Call the 
submit_task
tool:
  • model_id: 
    st-ai/super-seed2
  • parameters:
    • prompt: Video description prompt
    • functionMode: 
      first_last_frames
    • filePaths: Array of image URLs (0 images = text-to-video, 1 image = first-frame-to-video, 2 images = first-last-frame video)
    • ratio: Aspect ratio
    • duration: Integer duration
    • model: 
      seedance_2.0_fast
      or 
      seedance_2.0
</details> <details> <summary><b>Script Method</b></summary>
bash
python .cursor/skills/seedance2-api/scripts/seedance_api.py submit \
  --model "st-ai/super-seed2" \
  --params '{
    "prompt": "Camera transitions smoothly from first frame to last frame, with natural and fluid visuals",
    "functionMode": "first_last_frames",
    "filePaths": ["https://first-frame.png", "https://last-frame.png"],
    "ratio": "16:9",
    "duration": 5,
    "model": "seedance_2.0_fast"
  }'
</details>

Step 7: 轮询等待视频结果

Step 7: Poll for Video Results

视频生成约需 10 分钟。
<details> <summary><b>MCP 方式</b></summary>
轮询策略：
  1. 提交后告知用户"视频正在生成，预计 10 分钟"
  2. 首次 60 秒后调用 
    get_task
    查询
  3. 之后每 90 秒查询一次
  4. 每次查询后向用户报告状态
</details> <details> <summary><b>脚本方式</b></summary>
推荐：使用 poll 自动轮询（后台运行，间隔 30s，超时 600s）：
bash
python .cursor/skills/seedance2-api/scripts/seedance_api.py poll \
  --task-id "TASK_ID_HERE" --interval 30 --timeout 600
脚本会在 stderr 打印轮询进度，完成后在 stdout 输出 JSON 结果。
手动分次查询：
bash
python .cursor/skills/seedance2-api/scripts/seedance_api.py query \
  --task-id "TASK_ID_HERE"
</details>
状态说明：
  • pending
    → "排队中..."
  • processing
    → "生成中..."
  • completed
    → 提取视频 URL 并展示给用户
  • failed
    → 告知失败原因，建议调整提示词重试
Video generation takes approximately 10 minutes.
<details> <summary><b>MCP Method</b></summary>
Polling Strategy:
  1. After submission, inform the user: "Video is being generated, expected to take 10 minutes"
  2. First call 
    get_task
    to query after 60 seconds
  3. Then query every 90 seconds thereafter
  4. Report the status to the user after each query
</details> <details> <summary><b>Script Method</b></summary>
Recommended: Use auto polling (runs in background, 30s interval, 600s timeout):
bash
python .cursor/skills/seedance2-api/scripts/seedance_api.py poll \
  --task-id "TASK_ID_HERE" --interval 30 --timeout 600
The script will print polling progress to stderr, and output JSON results to stdout upon completion.
Manual Query in Batches:
bash
python .cursor/skills/seedance2-api/scripts/seedance_api.py query \
  --task-id "TASK_ID_HERE"
</details>
Status Explanations:
  • pending
    → "Queued..."
  • processing
    → "Generating..."
  • completed
    → Extract video URL and present to user
  • failed
    → Inform user of failure reason, suggest adjusting prompt and retrying

完整流程示例

Complete Workflow Example

用户说："帮我做一个宇航员在火星行走的视频"
User says: "Help me make a video of an astronaut walking on Mars"

MCP 可用时

When MCP is Available

1. 收集信息 → 15秒，16:9，电影级科幻风格，无现成素材

2. 用 Seedream 4.5 生成宇航员图 + 火星场景图
   submit_task("fal-ai/bytedance/seedream/v4.5/text-to-image", {...})
   → 轮询 get_task → 获得图片 URL

3. 编写提示词 → 提交视频任务
   submit_task("st-ai/super-seed2", {...})

4. 轮询 get_task，~10分钟后获得视频 URL
1. Collect information → 15 seconds, 16:9, cinematic sci-fi style, no existing materials

2. Generate astronaut image + Mars scene image using Seedream 4.5
   submit_task("fal-ai/bytedance/seedream/v4.5/text-to-image", {...})
   → Poll get_task → Obtain image URLs

3. Write prompt → Submit video task
   submit_task("st-ai/super-seed2", {...})

4. Poll get_task, obtain video URL after ~10 minutes

MCP 不可用时（脚本模式）

When MCP is Unavailable (Script Mode)

1. 收集信息 → 15秒，16:9，电影级科幻风格

2. 生成参考图：
   python scripts/seedance_api.py submit \
     --model "fal-ai/bytedance/seedream/v4.5/text-to-image" \
     --params '{"prompt":"An astronaut in white spacesuit on Mars...","image_size":"landscape_16_9"}'
   → 拿到 task_id

3. 轮询图片结果：
   python scripts/seedance_api.py poll --task-id "xxx" --interval 10 --timeout 180
   → 拿到图片 URL

4. 提交视频任务：
   python scripts/seedance_api.py submit \
     --model "st-ai/super-seed2" \
     --params '{"prompt":"...分镜提示词...","functionMode":"omni_reference","image_files":["图片URL"],"ratio":"16:9","duration":15,"model":"seedance_2.0_fast"}'
   → 拿到 task_id

5. 轮询视频结果：
   python scripts/seedance_api.py poll --task-id "xxx" --interval 30 --timeout 600
   → 拿到视频 URL
1. Collect information → 15 seconds, 16:9, cinematic sci-fi style

2. Generate reference images:
   python scripts/seedance_api.py submit \
     --model "fal-ai/bytedance/seedream/v4.5/text-to-image" \
     --params '{"prompt":"An astronaut in white spacesuit on Mars...","image_size":"landscape_16_9"}'
   → Get task_id

3. Poll for image results:
   python scripts/seedance_api.py poll --task-id "xxx" --interval 10 --timeout 180
   → Get image URLs

4. Submit video task:
   python scripts/seedance_api.py submit \
     --model "st-ai/super-seed2" \
     --params '{"prompt":"...storyboard prompt...","functionMode":"omni_reference","image_files":["image URL"],"ratio":"16:9","duration":15,"model":"seedance_2.0_fast"}'
   → Get task_id

5. Poll for video results:
   python scripts/seedance_api.py poll --task-id "xxx" --interval 30 --timeout 600
   → Get video URL

模型参数速查

Quick Reference for Model Parameters

Seedream 4.5 文生图

Seedream 4.5 Text-to-Image

参数类型必填说明
promptstring是图片描述
image_sizestring否auto_2K / auto_4K / square_hd / portrait_4_3 / portrait_16_9 / landscape_4_3 / landscape_16_9
num_imagesint否1-6，默认 1
ParameterTypeRequiredDescription
promptstringYesImage description
image_sizestringNoauto_2K / auto_4K / square_hd / portrait_4_3 / portrait_16_9 / landscape_4_3 / landscape_16_9
num_imagesintNo1-6, default 1

Seedream 4.5 图像编辑

Seedream 4.5 Image Editing

参数类型必填说明
promptstring是编辑指令，用 Figure 1/2/3 引用
image_urlsarray是输入图片 URL 列表
image_sizestring否同上
num_imagesint否1-6，默认 1
ParameterTypeRequiredDescription
promptstringYesEditing instructions, reference with Figure 1/2/3
image_urlsarrayYesList of input image URLs
image_sizestringNoSame as above
num_imagesintNo1-6, default 1

Seedance 2.0 视频（全能模式 omni_reference）

Seedance 2.0 Video (Omni-reference Mode 
omni_reference
)

参数类型必填说明
promptstring是分镜提示词，用 @image_file_N/@video_file_N/@audio_file_N 引用
functionModestring否
omni_reference
（默认）
image_filesarray否参考图片 URL 数组（最多 9 张）
video_filesarray否参考视频 URL 数组（最多 3 个，总时长 ≤ 15s）
audio_filesarray否参考音频 URL 数组（最多 3 个）
ratiostring否21:9 / 16:9 / 4:3 / 1:1 / 3:4 / 9:16
durationinteger否4-15 整数，默认 5
modelstring否seedance_2.0_fast（默认）/ seedance_2.0
ParameterTypeRequiredDescription
promptstringYesStoryboard prompt, reference with @image_file_N/@video_file_N/@audio_file_N
functionModestringNo
omni_reference
(default)
image_filesarrayNoArray of reference image URLs (max 9)
video_filesarrayNoArray of reference video URLs (max 3, total duration ≤ 15s)
audio_filesarrayNoArray of reference audio URLs (max 3)
ratiostringNo21:9 / 16:9 / 4:3 / 1:1 / 3:4 / 9:16
durationintegerNo4-15, default 5
modelstringNoseedance_2.0_fast (default) / seedance_2.0

Seedance 2.0 视频（首尾帧模式 first_last_frames）

Seedance 2.0 Video (First-Last Frame Mode 
first_last_frames
)

参数类型必填说明
promptstring是视频描述提示词
functionModestring是
first_last_frames
filePathsarray否图片 URL 数组（0张=文生视频，1张=首帧，2张=首尾帧）
ratiostring否21:9 / 16:9 / 4:3 / 1:1 / 3:4 / 9:16
durationinteger否4-15 整数，默认 5
modelstring否seedance_2.0_fast（默认）/ seedance_2.0
ParameterTypeRequiredDescription
promptstringYesVideo description prompt
functionModestringYes
first_last_frames
filePathsarrayNoArray of image URLs (0 = text-to-video, 1 = first frame, 2 = first-last frames)
ratiostringNo21:9 / 16:9 / 4:3 / 1:1 / 3:4 / 9:16
durationintegerNo4-15, default 5
modelstringNoseedance_2.0_fast (default) / seedance_2.0

工具速查

Quick Reference for Tools

MCP 工具

MCP Tools

操作工具关键参数
提交任务submit_taskmodel_id, parameters
查询结果get_tasktask_id
上传图片upload_imageimage_url 或 image_data
查询余额get_balance无
OperationToolKey Parameters
Submit Tasksubmit_taskmodel_id, parameters
Query Resultget_tasktask_id
Upload Imageupload_imageimage_url or image_data
Query Balanceget_balanceNone

脚本命令（MCP 不可用时）

Script Commands (When MCP is Unavailable)

操作命令说明
提交任务
python scripts/seedance_api.py submit --model MODEL --params '{...}'
返回 task_id
单次查询
python scripts/seedance_api.py query --task-id ID
返回当前状态
自动轮询
python scripts/seedance_api.py poll --task-id ID --interval N --timeout N
阻塞直到完成
查询余额
python scripts/seedance_api.py balance
返回账户余额
上传图片
python scripts/seedance_api.py upload --image-url URL
或 
--image-path PATH
返回图片 URL
脚本路径说明： 以上命令中的 
scripts/seedance_api.py
是相对于 
.cursor/skills/seedance2-api/
目录的路径。实际执行时使用完整路径 
.cursor/skills/seedance2-api/scripts/seedance_api.py
，或先 cd 到 skill 目录。
OperationCommandDescription
Submit Task
python scripts/seedance_api.py submit --model MODEL --params '{...}'
Returns task_id
Single Query
python scripts/seedance_api.py query --task-id ID
Returns current status
Auto Polling
python scripts/seedance_api.py poll --task-id ID --interval N --timeout N
Blocks until completion
Query Balance
python scripts/seedance_api.py balance
Returns account balance
Upload Image
python scripts/seedance_api.py upload --image-url URL
or 
--image-path PATH
Returns image URL
Script Path Note: The 
scripts/seedance_api.py
in the above commands is relative to the 
.cursor/skills/seedance2-api/
directory. When executing, use the full path 
.cursor/skills/seedance2-api/scripts/seedance_api.py
, or cd to the skill directory first.

Seedance 2.0 限制

Seedance 2.0 Limitations

  • 不支持上传写实真人脸部素材
  • 最多 12 个文件：图片 ≤ 9 + 视频 ≤ 3 + 音频 ≤ 3
  • 视频/音频参考总时长 ≤ 15 秒
  • 含视频参考会消耗更多积分
  • Does not support uploading realistic human face materials
  • Maximum 12 files: images ≤9 + videos ≤3 + audio ≤3
  • Total duration of video/audio references ≤15 seconds
  • Using video references consumes more credits

更多资源

Additional Resources

详细分镜模板、完整示例和镜头词汇表见 reference.md。
Detailed storyboard templates, complete examples, and camera vocabulary can be found in reference.md.