seedance-2.0-kz / seedance-2.0-kz-fast

This page only indicates the currently available model of the MoonNexAI waltz-2.0-kz series: seedance-2.0-kz-fast and seedance-2.0-kz. If you want the most complete version of the Seedance parameter, give priority to this page.

Compatibility statement: Old model names seedance-2.0-fast and seedance-2.0 are still available, and MoonNexAI will map this group of KZ public models; new systems should be filled in directly seedance-2.0-kz-fast or seedance-2.0-kz.

Include Models

Models	Annotations
`seedance-2.0-kz-fast`	Default fast-tracked version of the main thrust, suitable for high-frequency generation.
`seedance-2.0-kz`	Standard version, suitable for a higher quality and `1080P` scene.

Fit to scene

Default Quick Generate

If you just want to run through the Seedance workflow, start with seedance-2.0-kz-fast.

Higher quality and 1080 P

If you clearly need 1080P, or if you want to have a more stable quality of the image, try seedance-2.0-kz first.

Multi-modular Reference

This group is suitable for both reference images, reference video and reference audio, and for first and last frames controls. New access complex media assets recommends priority references[], and then map to specific media assets models by model capacity.

MoonNexAI Current Clear Support

Capacity	Support
text-to-video	Support
Single Figure image-to-video	Support
`references[]` Universal media assets	Support
First and last frames	Support
Multi-reference images	Supported, maximum `9`
reference video	Support, top `3`
reference audio	Supported, maximum `3` segment
`Asset://` references	Supported
`generate_audio`	Support
`watermark`	Support
`seed`	Support
`web_search`	Support
`super_resolution_config`	Support

ips This type of additional copyright release field is not currently covered by the MoonNexAI public file commitment. If you really want, contact MoonNexAI to confirm the model configuration.

Subbar for request Parameters

MoonNexAI Unified Universal Video Parameters

The common access is POST /v1/videos. The client uses stable prompt, references[], aspect_ratio and size field, and the line differences are mapd by the service.

field	Type	Annotations
`model`	string	`seedance-2.0-kz-fast` or `seedance-2.0-kz`.
`prompt`	string	Video description; use `@alias` to refer to entries in `references[]`.
`references[]`	array	Universal media assets arrays, entries containing `media_type`, `role`, `url` and optional `alias`.
`duration`	number	Target duration, completed by the selected model capability matrix.
`aspect_ratio`	string	Video aspect ratio.
`size`	string	Output size, e.g. `720p` or `1080p`.
`generate_audio`	boolean	Whether or not to generate an synchronous audio.
`watermark`	boolean	Add watermarks.
`seed`	integer	Fixed random seeds.
`callback_url`	string	Optional task callback URL.

Volcengine-compatible compatible parameters

The compatible endpoint is POST /api/v3/contents/generations/tasks. It accepts Volcengine-compatible content[], ratio and resolution. Do not send both references[] and content[] in one request.

field	Type	Annotations
`model`	string	The name of the MoonNexAI public model is still being filled in.
`content[]`	array	Official multi-modular content arrays, text entries using `type=text`, media entries using `image_url`, `video_url` or `audio_url`.
`duration`	number	Compatible duration field.
`ratio`	string	Compatible aspect-ratio field.
`resolution`	string	Compatible output-resolution field.
`generate_audio`	boolean	Official audio generation switch.
`watermark`	boolean	Official watermark switch.
`seed`	integer	Official random seeds.
`callback_url`	string	Official callback address.

Common request Modes

seedance-2.0-kz-fast and seedance-2.0-kz support the common MoonNexAI media format. Prefer references[] for new mixed-media integrations, with media_type, role, url and optional alias on each entry. Legacy images[], videos[], audios[], image, reference_image_urls, first_frame_url, last_frame_url, audio_url and content[] remain compatible.

Basic example (MoonNexAI UCP)

bash

curl https://moonnexai.com/v1/videos \
  -H "Authorization: Bearer <MOONNEXAI_API_KEY>" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "seedance-2.0-kz-fast",
    "prompt": "让 @hero 自然转身，镜头稳定推进，真实光照。",
    "references": [
      {
        "media_type": "image",
        "role": "reference_image",
        "url": "https://example.com/reference.png",
        "alias": "hero"
      }
    ],
    "duration": 5,
    "aspect_ratio": "16:9",
    "size": "720p"
  }'

Basic example (Volcengine-compatible compatible parameters)

bash

curl https://moonnexai.com/api/v3/contents/generations/tasks \
  -H "Authorization: Bearer <MOONNEXAI_API_KEY>" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "seedance-2.0-kz-fast",
    "content": [
      { "type": "text", "text": "让 @hero 自然转身，镜头稳定推进，真实光照。" },
      { "type": "image_url", "role": "reference_image", "image_url": { "url": "https://example.com/reference.png" }, "alias": "hero" }
    ],
    "duration": 5,
    "ratio": "16:9",
    "resolution": "720p"
  }'

Example first and last frames

json

{
  "model": "seedance-2.0-kz-fast",
  "prompt": "镜头从人物正面推进到人物转身后的背影。",
  "images": [
    {
      "url": "https://example.com/first.png",
      "role": "first_frame"
    },
    {
      "url": "https://example.com/last.png",
      "role": "last_frame"
    }
  ],
  "duration": 5,
  "ratio": "16:9",
  "resolution": "720p"
}

Corresponding Volcengine-compatible compatible:

json

{
  "model": "seedance-2.0-kz-fast",
  "content": [
    { "type": "text", "text": "镜头从人物正面推进到人物转身后的背影。" },
    { "type": "image_url", "role": "first_frame", "image_url": { "url": "https://example.com/first.png" }, "alias": "first" },
    { "type": "image_url", "role": "last_frame", "image_url": { "url": "https://example.com/last.png" }, "alias": "last" }
  ],
  "duration": 5,
  "ratio": "16:9",
  "resolution": "720p"
}

Multigraph + MultiVideo + Multi Audio Example (MoonNexAI Unified Generic Parameters)

The KZ series supports a maximum of 9 pictures, 3 videos and 3 segments of audio. Below is a smaller combination of 2 + 2 + 2 to demonstrate a uniform references[]:

json

{
  "model": "seedance-2.0-kz-fast",
  "prompt": "保持 @hero1、@hero2 的角色一致，参考 @motion1、@motion2 的运镜，并混合 @audio1、@audio2 的背景声音。",
  "references": [
    { "media_type": "image", "role": "reference_image", "url": "https://example.com/hero-1.png", "alias": "hero1" },
    { "media_type": "image", "role": "reference_image", "url": "https://example.com/hero-2.png", "alias": "hero2" },
    { "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
}

seedance-2.0-kz uses the same structure; request returns parameter error before creation when it exceeds the upper limit of a single class or total.

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

json

{
  "model": "seedance-2.0-kz-fast",
  "content": [
    { "type": "text", "text": "保持 @hero1、@hero2 的角色一致，参考 @motion1、@motion2 的运镜，并混合 @audio1、@audio2 的背景声音。" },
    { "type": "image_url", "role": "reference_image", "image_url": { "url": "https://example.com/hero-1.png" }, "alias": "hero1" },
    { "type": "image_url", "role": "reference_image", "image_url": { "url": "https://example.com/hero-2.png" }, "alias": "hero2" },
    { "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
}

Use uploaded assets

After upload and review, query asset details and confirm generation_ready=true and preparation_status=ready. Explicitly include Asset://asset_xxx in every video request. Setting only input_type: "reference" does not load assets automatically.

reference images video:

json

{
  "model": "seedance-2.0-kz-fast",
  "prompt": "图中女孩对着镜头说“茄子”，360度环绕运镜",
  "mode": "fast",
  "resolution": "720p",
  "ratio": "adaptive",
  "duration": 5,
  "generate_audio": true,
  "watermark": false,
  "web_search": false,
  "input_type": "reference",
  "generation_type": "video",
  "content": [
    {
      "type": "image_url",
      "role": "reference_image",
      "image_url": {
        "url": "Asset://asset_xxx"
      }
    }
  ]
}

First-frame image-to-video:

json

{
  "model": "seedance-2.0-kz-fast",
  "prompt": "图中女孩对着镜头说“茄子”，360度环绕运镜",
  "input_type": "reference",
  "first_frame_url": "Asset://asset_xxx",
  "duration": 5,
  "ratio": "adaptive",
  "resolution": "720p"
}

Superscore Example

json

{
  "model": "seedance-2.0-kz",
  "prompt": "城市夜景延时摄影，车流光轨清晰，画面稳定。",
  "duration": 5,
  "ratio": "16:9",
  "resolution": "1080p",
  "super_resolution_config": {
    "resolution": "4k",
    "scene": "aigc",
    "tool_version": "professional",
    "fps": 60
  }
}

Other field compatible

The following field is a historical or extended compatible writing method without altering the agreed boundary in the two columns above:

field	The agreement to which it belongs	Annotations
`images[]` / `videos[]` / `audios[]`	MoonNexAI Compatibility	Split media assets arrays by media type.
`image` / `image_url` / `video_url` / `audio_url`	MoonNexAI Compatibility	Single media quick field.
`reference_image_urls` / `first_frame_url` / `last_frame_url`	MoonNexAI-compatible	Shortcuts for multiple reference images and first/last frames.
`input_type`	MoonNexAI Compatibility	Common values are `reference`, `first_last_frame`.
`web_search` / `super_resolution_config`	MoonNexAI Extension	Networking enhancement and chain over-speculation.

Creation response fields

When task is created, save the task_id or id returned by MoonNexAI first. This task ID is used for follow-up queries.

field	Type	Annotations
`id`	string	MoonNexAI task ID. Usually the same as `task_id`.
`task_id`	string	MoonNexAI task ID. It is recommended to save this field for query.
`data.task_id`	string	Model task ID. Use only for supplementary search, not recommended as business primary key.
`status`	string	Optional field. The creation of response does not necessarily appear, and does not depend on it to judge whether task is complete.
`message`	string	API message. An empty string usually means that request is accepted.
`request_id`	string	request ID.
`trace_id`	string	Track ID.
`metadata.request_id`	string	request ID related to this request.
`metadata.trace_id`	string	Track ID related to this request.

Creates response with only task_id and no video URLs, which is normal. Continue searching for task results.

Query response fields

When searching task, the field phase varies. The client system recommends that it be read on the basis of the existence of field and do not assume that each response has the same top layer field as the other.

field	Type	Annotations
`id`	string	MoonNexAI task ID.
`task_id`	string	MoonNexAI task ID.
`status`	string	Optional. task status; part response may not have a top layer `status`.
`data.status`	string	Optional. task status. If the top layer `status` does not exist, first read this field.
`data.error`	string	Reason for failure. Read first on `data.status=failed`.
`data.duration`	number	Output duration in seconds; usually `0` while processing.
`data.usage`	object	Measuring information. `0` may be the case if not completed or failed.
`url`	string	Optional. Final video URL.
`video_url`	string	Optional. Final video URL.
`result_url`	string	Optional. Final video URL.
`output.url`	string	Optional. Final video URL.
`metadata.url`	string	Optional. Final video URL or content reading address.
`data.video_url`	string	Optional. Final video URL. Read first when you have finished.
`request_id`	string	request ID.
`trace_id`	string	Track ID.

Status judgement

Status reading suggestions in this order:

If the top layer status exists, read the top layer status first.
If the top layer status does not exist, read data.status.
If neither state field exists, but only task_id in response indicates that task has been created, please continue with the polling query.
Continue the query if the state is pending, submitted or running.
If the state is succeeded, read the video URL field.
If failed, read data.error and save task_id, request_id, trace_id.

Video URL Read Order

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

text

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

After the task succeeds and has no error, save an accessible video URL as the final result. If a successful response has no URL, query the same task_id again; inspect errors for failed tasks.

Query Results

It is recommended to save task_id, request, final video URL, request_id and trace_id in the operating system. When response does not have a top layer status, you can just read the order above.

Back Overview

Seedance-2 Video Overview

Conversation APIs

Embeddings

Audio

Models

GPT Image

Gemini Nano

Ideogram Image

Midjourney Task

Seedance-2 Video

Moon-2 Video

seedance-2.0-kz / seedance-2.0-kz-fast

Include Models

Fit to scene

Default Quick Generate

Higher quality and 1080 P

Multi-modular Reference

MoonNexAI Current Clear Support

Subbar for request Parameters

MoonNexAI Unified Universal Video Parameters

Volcengine-compatible compatible parameters

Common request Modes

Basic example (MoonNexAI UCP)

Basic example (Volcengine-compatible compatible parameters)

Example first and last frames

Multigraph + MultiVideo + Multi Audio Example (MoonNexAI Unified Generic Parameters)

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

Use uploaded assets

Superscore Example

Other field compatible

Creation response fields

Query response fields

Status judgement

Video URL Read Order

Query Results

Back Overview

Ideogram Image

Midjourney Task

Seedance-2 Video

Moon-2 Video

seedance-2.0-kz / seedance-2.0-kz-fast ​

Include Models ​

Fit to scene ​

Default Quick Generate ​

Higher quality and 1080 P ​

Multi-modular Reference ​

MoonNexAI Current Clear Support ​

Subbar for request Parameters ​

MoonNexAI Unified Universal Video Parameters ​

Volcengine-compatible compatible parameters ​

Common request Modes ​

Basic example (MoonNexAI UCP) ​

Basic example (Volcengine-compatible compatible parameters) ​

Example first and last frames ​

Multigraph + MultiVideo + Multi Audio Example (MoonNexAI Unified Generic Parameters) ​

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

Use uploaded assets ​

Superscore Example ​

Other field compatible ​

Creation response fields ​

Query response fields ​

Status judgement ​

Video URL Read Order ​

Query Results ​

Back Overview ​

seedance-2.0-kz / seedance-2.0-kz-fast

Include Models

Fit to scene

Default Quick Generate

Higher quality and 1080 P

Multi-modular Reference

MoonNexAI Current Clear Support

Subbar for request Parameters

MoonNexAI Unified Universal Video Parameters

Volcengine-compatible compatible parameters

Common request Modes

Basic example (MoonNexAI UCP)

Basic example (Volcengine-compatible compatible parameters)

Example first and last frames

Multigraph + MultiVideo + Multi Audio Example (MoonNexAI Unified Generic Parameters)

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

Use uploaded assets

Superscore Example

Other field compatible

Creation response fields

Query response fields

Status judgement

Video URL Read Order

Query Results

Back Overview