seedance-2.0-cl / seedance-2.0-cl-fast / seedance-2.0-cl-mini

seedance-2.0-cl, seedance-2.0-cl-fast and seedance-2.0-cl-mini are independent MoonNexAI Seedance-2 model lines for text-to-video, image-to-video, reference images and video references. Set model to the exact selected public ID.

This page only describes the CL series. media assets ID, uploading process and parameter boundaries are based on the three CL models of this page.

seedance-2.0-cl-b-fast and seedance-2.0-cl-b-mini are stand-alone CL-B model lines. For the official content[] compatibility range, see CL-B Series Page, do not directly apply the three CL types of capability boundaries on this page.

Current Configuration

Item	Annotations
Recommended entrance	`POST /v1/videos`
Query Entry	`GET /v1/videos/{task_id}` or `GET /v1/tasks/{task_id}`
Recommended Fast Model	`seedance-2.0-cl-fast`
Recommended high-quality models	`seedance-2.0-cl`
Recommended Light Model	`seedance-2.0-cl-mini`
Recommended reference fields	`image_url`, `video_url`, `reference_video_url`, `first_frame_url`, `references[]`, `content[]`
Recommended reference sources	Stable public image URLs; for video references, prepare a MoonNexAI `Asset://asset_xxx` for this model line through the asset API
Default Audio	`generate_audio` processing by `true` when omitted; visible `false` can turn off the generation of audio
Result URL	Read `url`, `video_url`, `result_url`, `output.url` or `metadata.url` after successful query

Capacity boundaries

Capacity	`seedance-2.0-cl-fast`	`seedance-2.0-cl`	`seedance-2.0-cl-mini`
text-to-video	Support	Support	Support
Single Figure image-to-video	Support	Support	Support
reference images Video	Support	Support	Support
reference video	Support	Support	Support
reference audio	Support	Support	Support
Generate Audio	Supported, default open	Supported, default open	Supported, default open
URL input for the network	Recommendations	Recommendations	Recommendations
MoonNexAI hosting URL	Support	Support	Support
`Asset://` references	Supported	Supported	Supported
Asset creation	Prepare `Asset://asset_xxx` for this model line through the asset API	Prepare `Asset://asset_xxx` for this model line through the asset API	Prepare `Asset://asset_xxx` for this model line through the asset API
Common resolution	`480P`, `720P`	`480P`, `720P`, `1080P`	`480P`, `720P`
duration	`4` - `15` seconds	`4` - `15` seconds	`4` - `15` seconds

Audio Switches

Three CL models are supported generate_audio:

Default to generate audio when generate_audio is omitted, with an equal price for the input "generate_audio": true.
Turns off the generated audio when the visible input "generate_audio": false is in, and the successful video does not contain a track.
generateAudio is accepted for compatibility; use generate_audio in new integrations.

Default access audio:

json

{
  "model": "seedance-2.0-cl-fast",
  "prompt": "A peaceful sunrise over a quiet lake with gentle ambient music.",
  "duration": 4,
  "aspect_ratio": "16:9",
  "size": "480p"
}

Turn off audio:

json

{
  "model": "seedance-2.0-cl-fast",
  "prompt": "A peaceful sunrise over a quiet lake.",
  "duration": 4,
  "aspect_ratio": "16:9",
  "size": "480p",
  "generate_audio": false
}

media assets policy

This model line supports the URL of the public network, MoonNexAI hosting URLs and Asset://asset_xxx media assets. The CL media assets preparation results are confirmed only on the actual CL model.

If your picture already has a stable public network URL, you can pass it directly in video-generated request.
If media assets is required, you can call /v1/assets/uploads, upload mode=asset and target models seedance-2.0-cl-fast, seedance-2.0-cl or seedance-2.0-cl-mini, and create Asset://asset_xxx available for this model line.
Prepare video references using type=video, purpose=reference_video and role=reference_video. Pass the resulting Asset://asset_xxx to video_url, reference_video_url or references[].url.
Please confirm that media assets details are generation_ready=true/preparation_status=ready before using Asset://asset_xxx.
Specify the actual CL model during asset creation and preparation, and check asset status before generation.
The client side can save the asset_xxx/Asset://asset_xxx of MoonNexAI and record which model line it uses to facilitate subsequent reuse and routing.

Create CL media assets Reference

bash

curl https://moonnexai.com/v1/assets/uploads \
  -H "Authorization: Bearer <MOONNEXAI_API_KEY>" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "seedance-2.0-cl-mini",
    "mode": "asset",
    "type": "image",
    "url": "https://example.com/product.png"
  }'

Create a CL video reference

bash

curl https://moonnexai.com/v1/assets/uploads \
  -H "Authorization: Bearer <MOONNEXAI_API_KEY>" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "seedance-2.0-cl",
    "mode": "asset",
    "type": "video",
    "purpose": "reference_video",
    "role": "reference_video",
    "url": "https://example.com/reference.mp4"
  }'

Subbar for request Parameters

MoonNexAI Unified Universal Video Parameters

The unified access is POST /v1/videos, and prompt, references[], aspect_ratio and size are recommended.

field	Type	Annotations
`model`	string	`seedance-2.0-cl-fast`, `seedance-2.0-cl` or `seedance-2.0-cl-mini`.
`prompt`	string	Video description and media assets binding instructions.
`references[]`	array	Generic media assets arrays, entries using `media_type`, `role`, `url` and optional `alias`.
`duration`	number	`4` - `15` sec.
`aspect_ratio`	string	Video aspect ratio.
`size`	string	`480p`, `720p` or sizes supported by the current model.
`generate_audio`	boolean	Whether or not to generate audio, default `true` when omitted.
`watermark`	boolean	Add watermarks.
`callback_url`	string	Optional callback address.

Volcengine-compatible compatible parameters

The POST /api/v3/contents/generations/tasks compatible entrance is used for official content[], ratio and resolution. The same request is not used for simultaneous references[] and content[].

field	Type	Annotations
`model`	string	Still fills out the name of the MoonNexAI public model for the CL series.
`content[]`	array	Official text, pictures, videos and audio entries.
`duration`	number	Compatible duration field; supports `4`–`15` seconds.
`ratio`	string	Compatible aspect-ratio field.
`resolution`	string	Compatible resolution field.
`generate_audio`	boolean	Official audio generation switch.
`watermark`	boolean	Official watermark switch.
`callback_url`	string	Official callback address.

Create a video task

Web Picture URL image-to-video

bash

curl https://moonnexai.com/v1/videos \
  -H "Authorization: Bearer <MOONNEXAI_API_KEY>" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "seedance-2.0-cl-fast",
    "prompt": "Animate this product image with a slow camera push-in and clean studio lighting.",
    "image_url": "https://example.com/product.png",
    "input_reference_role": "first_frame",
    "duration": 4,
    "aspect_ratio": "16:9",
    "size": "480p",
    "generate_audio": false,
    "watermark": false
  }'

Assembly://image-to-video

bash

curl https://moonnexai.com/v1/videos \
  -H "Authorization: Bearer <MOONNEXAI_API_KEY>" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "seedance-2.0-cl-fast",
    "prompt": "Animate the uploaded product with a premium studio camera move.",
    "image_url": "Asset://asset_xxx",
    "input_reference_role": "first_frame",
    "duration": 4,
    "aspect_ratio": "16:9",
    "size": "480p",
    "watermark": false
  }'

Please confirm that media assets is prepared by pressing seedance-2.0-cl-fast / seedance-2.0-cl and generation_ready=true, preparation_status=ready in media assets details, and Asset://asset_xxx is used to generate video of this model line.

reference video+ Picture URL

reference video can be created as Asset://asset_xxx first. Photo references can be sent directly to image_url if there is a stable public network URL:

bash

curl https://moonnexai.com/v1/videos \
  -H "Authorization: Bearer <MOONNEXAI_API_KEY>" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "seedance-2.0-cl",
    "prompt": "保留参考视频中的动作和镜头节奏，把人物上衣图案替换为参考图片风格。",
    "video_url": "Asset://asset_video_xxx",
    "image_url": "https://example.com/reference-shirt.png",
    "duration": 5,
  "aspect_ratio": "1:1",
  "size": "720p"
  }'

When submitted successfully, input_references in response usually shows pictures and videos bounded:

json

[
  {
    "alias": "图片1",
    "binding_status": "bound",
    "role": "reference_image",
    "type": "image"
  },
  {
    "alias": "视频1",
    "binding_status": "bound",
    "role": "reference_video",
    "type": "video"
  }
]

References[ ]

If your system is already in uniform, references[], you can also submit it in this way:

json

{
  "model": "seedance-2.0-cl-mini",
  "prompt": "Let @product slowly rotate on a clean studio table, cinematic lighting.",
  "references": [
    {
      "media_type": "image",
      "role": "first_frame",
      "url": "Asset://asset_xxx",
      "alias": "product"
    }
  ],
  "duration": 4,
  "aspect_ratio": "16:9",
  "size": "480p",
  "watermark": false
}

The seedance-2.0-cl-mini and CL series uses the same system of parameters; the difference is that Mini currently only uses 480p / 720p resolution. duration also supports 4 - 15 seconds.

Multitage + Multitelevision + Multi Audio Example

CL uses the same mixed-reference structure as Volcengine-compatible Seedance APIs. The example uses 2 images, 2 videos and 2 audio files. Current model capabilities determine per-type and total limits:

json

{
  "model": "seedance-2.0-cl",
  "prompt": "保持 @product1、@product2 的外观一致，参考 @motion1、@motion2 的动作节奏，并融合 @audio1、@audio2 的声音氛围。",
  "references": [
    { "media_type": "image", "role": "reference_image", "url": "https://example.com/product-1.png", "alias": "product1" },
    { "media_type": "image", "role": "reference_image", "url": "https://example.com/product-2.png", "alias": "product2" },
    { "media_type": "video", "role": "reference_video", "url": "https://example.com/motion-1.mp4", "alias": "motion1" },
    { "media_type": "video", "role": "reference_video", "url": "https://example.com/motion-2.mp4", "alias": "motion2" },
    { "media_type": "audio", "role": "reference_audio", "url": "https://example.com/audio-1.mp3", "alias": "audio1" },
    { "media_type": "audio", "role": "reference_audio", "url": "https://example.com/audio-2.mp3", "alias": "audio2" }
  ],
  "duration": 5,
  "aspect_ratio": "16:9",
  "size": "720p",
  "generate_audio": true
}

references[].media_type can be used on the CL page with image, video or audio. The fine differences of different models are configured by model capabilities for pre-creation verification, so that the client maintains the same set of structures.

Multigraph + MultiVideo + Multi Audio Example (Volcengine-compatible compatible parameters)

json

{
  "model": "seedance-2.0-cl",
  "content": [
    { "type": "text", "text": "保持 @product1、@product2 的外观一致，参考 @motion1、@motion2 的动作节奏，并融合 @audio1、@audio2。" },
    { "type": "image_url", "role": "reference_image", "image_url": { "url": "https://example.com/product-1.png" }, "alias": "product1" },
    { "type": "image_url", "role": "reference_image", "image_url": { "url": "https://example.com/product-2.png" }, "alias": "product2" },
    { "type": "video_url", "role": "reference_video", "video_url": { "url": "https://example.com/motion-1.mp4" }, "alias": "motion1" },
    { "type": "video_url", "role": "reference_video", "video_url": { "url": "https://example.com/motion-2.mp4" }, "alias": "motion2" },
    { "type": "audio_url", "role": "reference_audio", "audio_url": { "url": "https://example.com/audio-1.mp3" }, "alias": "audio1" },
    { "type": "audio_url", "role": "reference_audio", "audio_url": { "url": "https://example.com/audio-2.mp3" }, "alias": "audio2" }
  ],
  "duration": 5,
  "ratio": "16:9",
  "resolution": "720p",
  "generate_audio": true
}

... that you can write in the same text as in the previous text.

If you need a multimodular content[] format, you can import the picture as an image_url entry; replace the example address with /api/v3/contents/generations/tasks when using the Volcengine-compatible entry:

Pictures, videos and audio entries suggest visible role. Pictures are commonly used first_frame or reference_image, videos are commonly used reference_video. If role is missing, partial compatible circuits may return 400 by service-end parameter.

json

{
  "model": "seedance-2.0-cl-fast",
  "content": [
    {
      "type": "text",
      "text": "Animate the product with a slow camera push-in and soft studio lighting."
    },
    {
      "type": "image_url",
      "role": "first_frame",
      "image_url": {
        "url": "https://example.com/product.png"
      }
    }
  ],
  "duration": 4,
  "ratio": "16:9",
  "resolution": "480p",
  "watermark": false
}

List of Parameters

MoonNexAI Unified Universal field

field	Type	Annotations
`model`	string	CL series public model name.
`prompt`	string	Video description.
`references[]`	array	Universal media assets reference array.
`references[].media_type` / `role` / `url` / `alias`	mixed	Media type, use, stable URL or `Asset://`, aliases are available.
`duration`	number	`4` - `15` sec.
`aspect_ratio`	string	Video aspect ratio.
`size`	string	output size; Mini only supports `480p` / `720p`.
`generate_audio`	boolean	Whether or not to generate audio, default `true` when omitted.
`watermark`	boolean	Add watermarks.
`callback_url`	string	Optional callback address.

Volcengine-compatible compatible field

field	Type	Annotations
`model`	string	Still fills out the name of the MoonNexAI public model for the CL series.
`content[]`	array	Official multi-modular input arrays.
`image_url` / `video_url` / `reference_video_url`	string	Official single media field.
`first_frame_url` / `last_frame_url`	string	Compatible first/last-frame fields.
`duration`	number	Compatible duration field.
`ratio`	string	Compatible aspect-ratio field.
`resolution`	string	Compatible resolution field.
`generate_audio` / `generateAudio`	boolean	Official audio generation switches and compatible aliases.
`watermark`	boolean	Official watermark switch.
`callback_url`	string	Official callback address.

Query Results

Create response has been submitted only for task, save id or task_id and continue searching:

bash

curl https://moonnexai.com/v1/videos/{task_id} \
  -H "Authorization: Bearer <MOONNEXAI_API_KEY>"

After the success of task, it is recommended that the final video address be read in this order:

text

data.video_url
video_url
url
result_url
output.url
metadata.url

Please first determine whether status is succeeded. If task is still pending, submitted or running, continue with polling and task_id; if status=failed, read error.message and save task_id, request_id and trace_id for easy searching.

Price and billing

The CL series is displayed and settled according to the MoonNexAI Seedance-2 caliber token caliber. The specific price, video input price and grouping multiplier is based on the MoonNexAI rear price page, model list API and the actual settlement log.

seedance-2.0-cl-mini supports output-token billing for 480P / 720P, with separate policies for requests with and without video input. Video input includes video_url, reference_video_url, videos[], entries in references[] with media_type=video or role=reference_video, and Asset://asset_xxx video references. Image URLs or image asset references alone are not video input. Check the console and billing records for current prices.

FAQ

CL media assets ID can be used across models?

media assets should be created or confirmed according to the actual generation model. The operating system recommends that asset_id, asset_ref, target CL model and media assets state be preserved simultaneously.

Is an upload of media assets automatically used for the next video generation?

No. Each time task is created, the image URL/ Asset://asset_xxx of image_url, first_frame_url, references[] or content[] must be uploaded in a visible manner. Only input_type will not automatically load the latest upload or recently approved media assets.

Which field should be saved by the URL?

First judge status=succeeded, then saves the first video address that is accessible in the URL reading order of this page. Do not treat URLs as final in pending, running or failed.

Conversation APIs

Embeddings

Audio

Models

GPT Image

Gemini Nano

Ideogram Image

Midjourney Task

Seedance-2 Video

Moon-2 Video

seedance-2.0-cl / seedance-2.0-cl-fast / seedance-2.0-cl-mini

Current Configuration

Capacity boundaries

Audio Switches

media assets policy

Create CL media assets Reference

Create a CL video reference

Subbar for request Parameters

MoonNexAI Unified Universal Video Parameters

Volcengine-compatible compatible parameters

Create a video task

Web Picture URL image-to-video

Assembly://image-to-video

reference video+ Picture URL

References[ ]

Multitage + Multitelevision + Multi Audio Example

Multigraph + MultiVideo + Multi Audio Example (Volcengine-compatible compatible parameters)

... that you can write in the same text as in the previous text.

List of Parameters

MoonNexAI Unified Universal field

Volcengine-compatible compatible field

Query Results

Price and billing

FAQ

CL media assets ID can be used across models?

Is an upload of media assets automatically used for the next video generation?

Which field should be saved by the URL?

Relevant Pages

Ideogram Image

Midjourney Task

Seedance-2 Video

Moon-2 Video

seedance-2.0-cl / seedance-2.0-cl-fast / seedance-2.0-cl-mini ​

Current Configuration ​

Capacity boundaries ​

Audio Switches ​

media assets policy ​

Create CL media assets Reference ​

Create a CL video reference ​

Subbar for request Parameters ​

MoonNexAI Unified Universal Video Parameters ​

Volcengine-compatible compatible parameters ​

Create a video task ​

Web Picture URL image-to-video ​

Assembly://image-to-video ​

reference video+ Picture URL ​

References[ ] ​

Multitage + Multitelevision + Multi Audio Example ​

Multigraph + MultiVideo + Multi Audio Example (Volcengine-compatible compatible parameters) ​

... that you can write in the same text as in the previous text. ​

List of Parameters ​

MoonNexAI Unified Universal field ​

Volcengine-compatible compatible field ​

Query Results ​

Price and billing ​

FAQ ​

CL media assets ID can be used across models? ​

Is an upload of media assets automatically used for the next video generation? ​

Which field should be saved by the URL? ​

Relevant Pages ​

seedance-2.0-cl / seedance-2.0-cl-fast / seedance-2.0-cl-mini

Current Configuration

Capacity boundaries

Audio Switches

media assets policy

Create CL media assets Reference

Create a CL video reference

Subbar for request Parameters

MoonNexAI Unified Universal Video Parameters

Volcengine-compatible compatible parameters

Create a video task

Web Picture URL image-to-video

Assembly://image-to-video

reference video+ Picture URL

References[ ]

Multitage + Multitelevision + Multi Audio Example

Multigraph + MultiVideo + Multi Audio Example (Volcengine-compatible compatible parameters)

... that you can write in the same text as in the previous text.

List of Parameters

MoonNexAI Unified Universal field

Volcengine-compatible compatible field

Query Results

Price and billing

FAQ

CL media assets ID can be used across models?

Is an upload of media assets automatically used for the next video generation?

Which field should be saved by the URL?

Relevant Pages