Moon-2.0-wc series

The moon-2.0-wc family provides 720P MoonNexAI Moon-2 video. Public models are moon-2.0-wc-720p, moon-2.0-wc-b-720p, moon-2.0-wc-b-720p-fast, moon-2.0-wc-c-720p, moon-2.0-wc-d-720p, moon-2.0-wc-e-720p and moon-2.0-wc-f-720p. Reference limits differ; consult this page's matrix before submitting.

When creating task, all seven models are written using references[] media assets. moon-2.0-wc-c-720p, moon-2.0-wc-d-720p, moon-2.0-wc-e-720p and moon-2.0-wc-f-720p are four separate models: C for up to 9 reference images, D/ E for a mix of 4 diagram + 3 video + 1 audio, F for up to 9 pictures, 3 videos and 3 audio, for a total of not more than 12 media assets audio images.

WC-C, WC-D, WC-E and WC-F are now open for call. duration indicates that the target is duration; the final piece of duration may vary with the content of task. When the operation requires strict duration, read and verify the actual media duration after task has succeeded.

Current Configuration

Item	Annotations
Recommended entrance	`POST /v1/videos`
Query Entry	`GET /v1/videos/{task_id}` or `GET /v1/tasks/{task_id}`
Recommended media field	`references[]`
Recommended media source	Stable URLs from your own storage for production use; `mode=oss` can provide temporary URLs for local integration tests
Output	720P
Standard Model duration	`moon-2.0-wc-720p` supports `4` to `14` seconds; compatible with `15` seconds and generated by `14` seconds
C / D / E / F duration	`moon-2.0-wc-c-720p`, `moon-2.0-wc-d-720p`, `moon-2.0-wc-e-720p`, `moon-2.0-wc-f-720p` accept `4` to `15` second target duration
C/ D / E aspect ratio	`16:9`, `9:16`, `1:1`, `4:3`, `3:4`
billing Method	Billed per task; actual price based on MoonNexAI console and billing records

Available Models

Models	Positioning	Annotations
`moon-2.0-wc-720p`	Standard 720P	The WC series base 720P multi-media assets model supports 4 to 14 seconds; 14 seconds for 15 seconds.
`moon-2.0-wc-b-720p`	Stand by 720 P	Supports 4 to 15 seconds and 9 diagram + 3 video + 3 audio references, not text-only.
`moon-2.0-wc-b-720p-fast`	720 P Fast	Supports 4 to 15 seconds and 9 diagram + 3 video + 3 audio references, not text-only.
`moon-2.0-wc-c-720p`	720P with up to 9 images	Text-only or up to 9 reference images; no video or audio references.
`moon-2.0-wc-d-720p`	4-3-1 Mixed reference 720 P	Supports pure text, or up to 4 pictures, 3 videos and 1 audio; audio must be accompanied by at least one picture or video.
`moon-2.0-wc-e-720p`	WC-E 4-3-1 Mixed reference 720P	Supports pure text, or up to 4 pictures, 3 videos and 1 audio; audio must be accompanied by at least one picture or video.
`moon-2.0-wc-f-720p`	WC-F Mixed Reference 720P	Supports pure text, or up to 9 pictures, 3 videos and 3 audio, for a total of up to 12 media assets; audio must be accompanied by at least one picture.

WC Model Capability Matrix

Models	Plain Text	Picture	Video	Audio	Total media assets	duration	Man.
`moon-2.0-wc-720p`	Support	Nine.	3	3	12	`4`-`14` seconds; compatible `15` seconds and generated by `14` seconds	Support
`moon-2.0-wc-b-720p`	Not supported	Nine.	3	3	12	`4`-`15` seconds	Support
`moon-2.0-wc-b-720p-fast`	Not supported	Nine.	3	3	12	`4`-`15` seconds	Support
`moon-2.0-wc-c-720p`	Support	Nine.	Not supported	Not supported	9	`4`-`15` second target duration	Support
`moon-2.0-wc-d-720p`	Support	4 sheets	3	1	8	`4`-`15` second target duration	Support
`moon-2.0-wc-e-720p`	Support	4 sheets	3	1	8	`4`-`15` second target duration	Support
`moon-2.0-wc-f-720p`	Support	Nine.	3	3	12	`4`-`15` second target duration	Support

D/ E/ F does not accept request, which is audio-only. D / E must submit at least one picture or video at the same time when using audio; F must submit at least one picture at the same time when using audio.

media assets Enter Border

Capacity	Support
`references[]` Universal media assets	Support
`Asset://` references	Not a current WC input method; use public URLs
Independent Audio media assets	Not recommended for individual use; please provide both pictures or video references by corresponding model, WC-F must provide photo references

The seven models are based on the “WC Model Capability Matrix” for pure text, media assets quantity and duration boundary.

C / D / E / F request Example

WC-C:9 Inline or plain text

json

{
  "model": "moon-2.0-wc-c-720p",
  "prompt": "参考 @产品 的外观，生成干净的棚拍产品广告。",
  "duration": 8,
  "ratio": "16:9",
  "resolution": "720p",
  "references": [
    {
      "media_type": "image",
      "role": "reference_image",
      "url": "https://example.com/product.png",
      "alias": "产品"
    }
  ]
}

WC-C's 9th photo is still available; the 10th picture, any video or any audio is rejected. The plain text request can be omitted references.

WC-D:4 Figure + 3 Video + 1 Audio

json

{
  "model": "moon-2.0-wc-d-720p",
  "prompt": "参考图片主体与视频动作节奏，并使用 @音乐 作为背景音乐。",
  "duration": 15,
  "ratio": "9:16",
  "resolution": "720p",
  "references": [
    {
      "media_type": "image",
      "role": "reference_image",
      "url": "https://example.com/subject.png",
      "alias": "主体"
    },
    {
      "media_type": "video",
      "role": "reference_video",
      "url": "https://example.com/motion.mp4",
      "alias": "动作"
    },
    {
      "media_type": "audio",
      "role": "reference_audio",
      "url": "https://example.com/music.mp3",
      "alias": "音乐"
    }
  ]
}

WC-D borders are 4 pictures, 3 videos, 1 audio and a total of 8 media assets. The fifth picture, 4 video, 2 audio and pure audio request are rejected. The pure text request can be omitted from references.

WC-E:4 Chart + 3 Video + 1 Audio

json

{
  "model": "moon-2.0-wc-e-720p",
  "prompt": "保持 @主体 的外观，参考 @动作 的运动节奏，并使用 @音乐 作为背景音乐。",
  "duration": 15,
  "ratio": "9:16",
  "resolution": "720p",
  "references": [
    {
      "media_type": "image",
      "role": "reference_image",
      "url": "https://example.com/subject.png",
      "alias": "主体"
    },
    {
      "media_type": "video",
      "role": "reference_video",
      "url": "https://example.com/motion.mp4",
      "alias": "动作"
    },
    {
      "media_type": "audio",
      "role": "reference_audio",
      "url": "https://example.com/music.mp3",
      "alias": "音乐"
    }
  ]
}

The media assets boundary of WC-E is the same as that of WC-D: up to 4 pictures, 3 videos, 1 audio and a total of 8 media assets, which must be accompanied by at least one picture or video. WC-E is an independent open model and must be called with an exact model name moon-2.0-wc-e-720p.

WC-F: Pure text or hybrid reference

request plain text can be omitted references:

json

{
  "model": "moon-2.0-wc-f-720p",
  "prompt": "一位舞者在现代舞台上完成流畅的独舞，镜头缓慢推进，灯光自然变化。",
  "duration": 15,
  "ratio": "9:16",
  "resolution": "720p"
}

When using reference media assets, the maximum number of pictures is 9 and the maximum number of videos is 3 and the maximum number of audio is 3 and the total number of references[] is 12. Audio must be matched with at least one picture, and only video cannot meet this condition:

json

{
  "model": "moon-2.0-wc-f-720p",
  "prompt": "保持 @主体 的外观，参考 @动作 的镜头节奏，并使用 @音乐 作为背景音乐。",
  "duration": 15,
  "ratio": "9:16",
  "resolution": "720p",
  "references": [
    {
      "media_type": "image",
      "role": "reference_image",
      "url": "https://example.com/subject.png",
      "alias": "主体"
    },
    {
      "media_type": "video",
      "role": "reference_video",
      "url": "https://example.com/motion.mp4",
      "alias": "动作"
    },
    {
      "media_type": "audio",
      "role": "reference_audio",
      "url": "https://example.com/music.mp3",
      "alias": "音乐"
    }
  ]
}

WC-F supports real-person content. Audio requires at least one image; audio alone or with only video is not accepted. Limits are 9 images, 3 videos and 3 audio files, with at most 12 references total, for example 6 images + 3 videos + 3 audio files. A tenth image, fourth video, fourth audio file or total above 12 is rejected.

Recommended Process

Organize mixed-media tasks using public URLs:

Images, videos and audio URLs that are intended to be accessed by MoonNexAI.
When you connect local media assets, you can call /v1/assets/uploads and use mode=oss to get a temporary test URL; for official business, use the stable public network URL that the customer has stored.
Call /v1/videos, setting references[].url to each public URL.

The WC series is currently accessible to the public network URL reference media assets. Do not use Asset://asset_xxx as the main input format for WC task.

moon-2.0-wc-720p, moon-2.0-wc-b-720p, moon-2.0-wc-b-720p-fast and moon-2.0-wc-f-720p accept at most 12 entries in references[]: up to 9 images, 3 videos and 3 audio files. WC-F requires an image when audio is used. C, D and E have separate limits in the WC capability matrix.

Prepare public network URL

WC series videos, pictures and audio references recommend the use of the MoonNexAI direct downloadable public network line. To reduce third-party media assets capture failure or media type miscalculation, it is recommended that:

URL path retains the real file suffix, e. g. .png, .jpg, .mp4, .wav, .mp3.
URL does not rely on login, Cookie, extra authentication Header, one-time download page or security chain verification.
Use of a straight chain of stable, short, uncomplicated query in the customer ' s own storage; mode=oss is used only as a temporary test capability for the association of local files and does not guarantee long-term validity.
When you use a sound as a background music, you clearly say " Use @music as background music" in prompt and do not write " voice, verb or lines " .

Local File Transfer OSS Test URL

bash

curl https://moonnexai.com/v1/assets/uploads \
  -H "Authorization: Bearer <MOONNEXAI_API_KEY>" \
  -F "mode=oss" \
  -F "model=moon-2.0-wc-720p" \
  -F "type=image" \
  -F "file=@reference.png"

After uploading successfully, enter the returned public web url/ public_url into the references[].url of Video task.

URL already available

bash

curl https://moonnexai.com/v1/videos \
  -H "Authorization: Bearer <MOONNEXAI_API_KEY>" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "moon-2.0-wc-720p",
    "prompt": "参考 @hero 的人物，在自己的房间里跳一段可爱的舞蹈。",
    "duration": 14,
    "ratio": "9:16",
    "resolution": "720p",
    "references": [
      {
        "media_type": "image",
        "role": "reference_image",
        "url": "https://example.com/reference.png",
        "alias": "hero"
      }
    ]
  }'

Create a video task

Single Chart Reference

bash

curl https://moonnexai.com/v1/videos \
  -H "Authorization: Bearer <MOONNEXAI_API_KEY>" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "moon-2.0-wc-720p",
    "prompt": "参考 @hero 的人物，在自己的房间里跳一段可爱的舞蹈。",
    "duration": 14,
    "ratio": "9:16",
    "resolution": "720p",
    "references": [
      {
        "media_type": "image",
        "role": "reference_image",
        "url": "https://example.com/reference.png",
        "alias": "hero"
      }
    ]
  }'

Multi-media assets grouping

json

{
  "model": "moon-2.0-wc-720p",
  "prompt": "让 @角色 在自己的房间里跳可爱的舞蹈，参考 @动作视频 的动作节奏，并使用 @音乐 作为背景音乐。",
  "duration": 14,
  "ratio": "9:16",
  "resolution": "720p",
  "references": [
    {
      "media_type": "image",
      "role": "character_reference",
      "url": "https://example.com/character.png",
      "alias": "角色"
    },
    {
      "media_type": "video",
      "role": "reference_video",
      "url": "https://example.com/motion.mp4",
      "alias": "动作视频"
    },
    {
      "media_type": "audio",
      "role": "reference_audio",
      "url": "https://example.com/music.mp3",
      "alias": "音乐"
    }
  ]
}

A mixed-media task accepts at most 12 references: up to 9 images, 3 videos and 3 audio files, subject to the combined limit. For example, use 6 images + 3 videos + 3 audio files. Omit @ from alias, do not bind one alias to different media in the same request, and use @alias in the prompt.

Picture + Background Music

json

{
  "model": "moon-2.0-wc-720p",
  "prompt": "参考 @人物 生成一段唯美人像视频。人物安静看向镜头、轻微微笑和自然摆动，不要开口讲话，不要说台词。请使用 @音乐 作为背景音乐，保留该音频的旋律、节奏和氛围。",
  "duration": 14,
  "ratio": "9:16",
  "resolution": "720p",
  "references": [
    {
      "media_type": "image",
      "role": "reference_image",
      "url": "https://example.com/person.png",
      "alias": "人物"
    },
    {
      "media_type": "audio",
      "role": "reference_audio",
      "url": "https://example.com/background-music.wav",
      "alias": "音乐"
    }
  ]
}

When references[].alias is omitted, the system is next to the name of the media type in order of 图片1, 视频1, 音频1. prompt matches @图片1 and @图片 1, 视频N, 音频N; numbers can then continue directly with the body, such as @图片1保持团扇. To be readable, it is still recommended to add spaces or points after aliases.

Custom aliases must be separated from subsequent text. For example, use @主体走到龙椅旁, not @主体走到龙椅旁. An unknown alias returns invalid_reference_alias; one alias bound to multiple different references returns duplicate_reference_alias. Binding helps identify the intended reference but does not guarantee distinct identities for multiple people. Also specify each person's position, clothing, props and actions.

Common field

field	Type	Annotations
`model`	string	Fill in any of the public model names in the " Available Models " page.
`prompt`	string	Video description. It is recommended to write about the subject, the action, the scene, the lens and the media assets binding.
`duration`	number/string	Target generates duration. C / D / E / F accepts `4` to `15` seconds; ultimately, media duration should be based on a successful outcome. Other models are available.
`ratio` / `aspect_ratio`	string	C/ D / E supports `16:9`, `9:16`, `1:1`, `4:3`, `3:4`.
`resolution`	string	It is recommended that `720p` be completed.
`references`	array	Recommended. Generic media assets reference array, single request totals up to 12.
`references[].media_type`	string	`image`, `video`, `audio` or `music`.
`references[].role`	string	Common values are `reference_image`, `character_reference`, `style_reference`, `reference_video`, `reference_audio`, `background_music`.
`references[].url`	string	URL or MoonNexAI hosting web URL.
`references[].alias`	string	Optional. Without `@`, the same request should be unique. Automatic `图片N`, `视频N`, `音频N` by media type is automatically generated when omitted.
`callback_url`	string	Optional. callback addresses after task is completed.

Query Results

Create response has only been submitted for task, save task_id and continue with the query:

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 the same task_id.

Attention

The WC series currently uses the web URL reference media assets; do not enter Asset://asset_xxx as WC task. moon-2.0-wc-b-720p and moon-2.0-wc-b-720p-fast follow the same rule.
moon-2.0-wc-720p supports 4 to 14 seconds; 15 seconds request is generated by 14 seconds for compatibility with an existing client. This rule does not affect the WC-B model.
The public video results only return to MoonNexAI's own controlled address; do not spell the URL of the result, use the full API return address.
references[] of base, B, B Fast and F models totals up to 12 reference documents; C/D/ E uses their respective stand-alone ceilings. task will be rejected when the corresponding model ceilings are exceeded.
Multi-sound or multi-video combinations suggest first-by-case confirmation of URLs that are publicly accessible and then submit task.
When reference audio is used as background music, please specify in prompt that “use @music as background music”; do not ask it to speak as a character voice or the content of the line.
If reference audio, reference video, is not identified in a stable way, check if URLs are directly downloadable public media files, with a real suffix and no additional authentication requirements.
If media assets security check fails or URLs cannot be accessed, replace with a stable URL that is anonymously accessible in the client's own storage; mode=oss can only be used for in-link searches.

Conversation APIs

Embeddings

Audio

Models

GPT Image

Gemini Nano

Ideogram Image

Midjourney Task

Seedance-2 Video

Moon-2 Video

Moon-2.0-wc series

Current Configuration

Available Models

WC Model Capability Matrix

media assets Enter Border

C / D / E / F request Example

WC-C:9 Inline or plain text

WC-D:4 Figure + 3 Video + 1 Audio

WC-E:4 Chart + 3 Video + 1 Audio

WC-F: Pure text or hybrid reference

Recommended Process

Prepare public network URL

Local File Transfer OSS Test URL

URL already available

Create a video task

Single Chart Reference

Multi-media assets grouping

Picture + Background Music

Common field

Query Results

Attention

Relevant Pages

Ideogram Image

Midjourney Task

Seedance-2 Video

Moon-2 Video

Moon-2.0-wc series ​

Current Configuration ​

Available Models ​

WC Model Capability Matrix ​

media assets Enter Border ​

C / D / E / F request Example ​

WC-C:9 Inline or plain text ​

WC-D:4 Figure + 3 Video + 1 Audio ​

WC-E:4 Chart + 3 Video + 1 Audio ​

WC-F: Pure text or hybrid reference ​

Recommended Process ​

Prepare public network URL ​

Local File Transfer OSS Test URL ​

URL already available ​

Create a video task ​

Single Chart Reference ​

Multi-media assets grouping ​

Picture + Background Music ​

Common field ​

Query Results ​

Attention ​

Relevant Pages ​

Moon-2.0-wc series

Current Configuration

Available Models

WC Model Capability Matrix

media assets Enter Border

C / D / E / F request Example

WC-C:9 Inline or plain text

WC-D:4 Figure + 3 Video + 1 Audio

WC-E:4 Chart + 3 Video + 1 Audio

WC-F: Pure text or hybrid reference

Recommended Process

Prepare public network URL

Local File Transfer OSS Test URL

URL already available

Create a video task

Single Chart Reference

Multi-media assets grouping

Picture + Background Music

Common field

Query Results

Attention

Relevant Pages