Seedance Asset Management

Seedance Asset Management lets you upload and organize media assets (images, videos, audio) for use with Seedance video generation. Once uploaded, assets are referenced by ID (asset://<ID>) in Seedance requests instead of public URLs.

Asset Management is independent of video generation. Use volc-asset, volc-asset-video, or volc-asset-audio when creating assets, then pass the resulting asset://<ID> to a supported Seedance model.

For specific content requests, select the corresponding model under the Direct resource group: volc-asset-nsfw for images, volc-asset-video-nsfw for videos, or volc-asset-audio-nsfw for audio. Each NSFW model ID adds the -nsfw suffix to the standard model ID.

Supported Seedance models

Official model	AnyFast model ID
Seedance 2.5	`seedance-2.5`
Seedance 2.0	`seedance-2.0`
Seedance 2.0 Fast	`seedance-2.0-fast`
Seedance 2.0 Mini	`seedance-2.0-mini`

Why use Asset Management?

Persistent storage — Assets are stored on Volcengine and don’t expire like temporary URLs
Multi-format support — Upload images, videos, and audio files
Organized by groups — Group related assets (e.g. per project or campaign)
Data isolation — Each API token sees only its own assets
Direct integration — Use asset://<ID> directly in Seedance image_url, video_url, audio_url fields

Supported formats

Type	Formats	Limits
Image	JPG, PNG, GIF, WebP, BMP, TIFF, HEIC, HEIF	< 30 MB, 300–6000 px, aspect ratio 0.4–2.5
Video	MP4, MOV	< 50 MB, 2–15s, 480p/720p, 24–60 FPS
Audio	MP3, WAV	< 15 MB, 2–15s

Seedance 2.5 enhanced input capabilities

This section applies only to seedance-2.5 and supplements the existing asset-management instructions above. Media inputs can be supplied through public URLs, supported Base64 data, or asset://<ID> references.

These limits describe media used as input to Seedance 2.5. For the upstream specification and examples, see the official Seedance 2.5 reference.

Image requirements

Input methods: Image URL, Base64-encoded image, or asset ID.
Formats: JPEG, PNG, WebP, BMP, TIFF, GIF, HEIC, and HEIF.
Dimensions: Aspect ratio (width/height) from 0.4 to 2.5; width and height from 300 to 6000 px.
Size: Each image must be smaller than 30 MB. The complete request body must not exceed 64 MB. Do not use Base64 for large files.

Generation mode	Number of images
Image-to-video with a first frame	1
Image-to-video with first and last frames	2
Multimodal reference-to-video	1–30

Video requirements

Input methods: Video URL or asset ID. Video Base64 input is not supported.
Formats: MP4 and MOV.
Resolution: 480p or 720p.
Duration and count: Each video must be 2–30 seconds. You can provide up to 10 reference videos, with a combined duration of no more than 30 seconds.
Dimensions: Aspect ratio (width/height) from 0.4 to 2.5; width and height from 300 to 6000 px.
Total pixels: Width multiplied by height must be between 409600 (640 × 640) and 8295044 (3326 × 2494).
Size: Each video must not exceed 200 MB.
Frame rate: 24–60 FPS.

Container	Common extension	MIME type	Supported codecs
MP4	`.mp4`	`video/mp4`	Video: H.264/AVC, H.265/HEVC; audio: AAC, MP3
QuickTime	`.mov`	`video/quicktime`	Video: H.264/AVC, H.265/HEVC; audio: AAC, MP3

Audio requirements

Input methods: Audio URL, Base64-encoded audio, or asset ID.
Formats: WAV and MP3.
Duration and count: Each audio clip must be 2–30 seconds. You can provide up to 10 reference audio clips, with a combined duration of no more than 30 seconds.
Size: Each audio clip must not exceed 15 MB. The complete request body must not exceed 64 MB. Do not use Base64 for large files.

Retention periods

Task records: Retained for 7 days. The query interval is [T-7 days, T), where T is the request time as a UTC Unix timestamp in seconds.
Generated video URLs: Retained for 24 hours and limited to 100 downloads. Download or transfer generated videos promptly.

The 24-hour period applies to generated video result URLs. It is separate from the temporary URL returned by GetAsset; continue to use asset://<asset_id> as the persistent asset reference.

Rate limits

RPM: The maximum number of video generation tasks that an account can create per minute for the same model. Requests above this limit return an error.
Concurrency: The maximum number of tasks that can be processed simultaneously for the same model under an account. Additional tasks wait in the queue.

Billing models

Model	Asset type	Description
`volc-asset`	Image	Default — no need to specify for images
`volc-asset-video`	Video	Required when uploading video
`volc-asset-audio`	Audio	Required when uploading audio

Asset review and activation time

Model	Asset type	Typical activation time
`volc-asset`	Image	6–9 seconds
`volc-asset-video`	Video	9–12 seconds
`volc-asset-audio`	Audio	3–6 seconds

These times are ideal estimates. Actual timing may vary based on file size, media format, and review queue load. When the review queue is congested, asset activation may be delayed by about 10 seconds to 2 minutes. Use the asset through asset://<ID> only after its status becomes Active.

Asynchronous processing and asset URLs

CreateAsset is asynchronous. The API returns an asset ID after accepting the upload request, but the asset still needs upstream preprocessing before it can be used for inference. Poll GetAsset until Status becomes Active.

Status	Meaning	Can be used in Seedance requests?
`Processing`	The asset is queued or being preprocessed.	No
`Active`	The asset is ready.	Yes, use `asset://<asset_id>`
`Failed`	Preprocessing failed.	No

Do not use or store the GetAsset.URL value as the long-term asset reference. Official examples mark this URL as valid for 12 hours. Use asset://<asset_id> in Seedance generation requests.

Workflow

Create an Asset Group        →  Group ID
Create an Asset in the Group →  Asset ID
Generate a video with Asset  →  async task ID
Poll for the result          →  pre-signed download URL

LivenessFace workflow

Use the LivenessFace flow when you need a private real-person portrait group. AnyFast creates a mobile verification page, the user completes liveness face verification on a phone, and the result becomes a LivenessFace asset group.

For the complete authorization flow and official terminology, see Real-human asset verification.

LivenessFace verification requires an API token created with the Byteplus-Direct group. Tokens routed to the regular AIGC asset group only support GroupType: "AIGC" and cannot create liveness sessions or LivenessFace assets.

Create a validation session

Call CreateVisualValidateSession to obtain H5Link and BytedToken.

cURL

curl https://www.anyfast.ai/volc/asset/CreateVisualValidateSession \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "CallbackURL": "https://example.com/callback"
  }'

Open the H5Link on a phone

Send the returned H5Link to the user and ask them to complete liveness face verification.

Query the result

After verification, call GetVisualValidateResult with the BytedToken. A successful verification returns a GroupId.

cURL

curl https://www.anyfast.ai/volc/asset/GetVisualValidateResult \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "BytedToken": "2026070222152680501D985EA34A3BE3D8"
  }'

Manage the LivenessFace group

Use the returned GroupId with UpdateAssetGroup, ListAssetGroups, CreateAsset, ListAssets, and GetAsset. When listing liveness groups or assets, include Filter.GroupType: "LivenessFace".

If verification is not completed or did not create a group, GetVisualValidateResult may return {"GroupId": ""}. Invalid or expired tokens return an upstream error.

When uploading an image to a LivenessFace group, the image must match the verified person. Mismatched faces return FaceMismatch and the asset status becomes Failed.

Step 1: Create an Asset Group

First, create a Group to obtain a Group ID.

curl https://www.anyfast.ai/volc/asset/CreateAssetGroup \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "volc-asset",
    "Name": "your-custom-name"
  }'

import requests

response = requests.post(
    "https://www.anyfast.ai/volc/asset/CreateAssetGroup",
    headers={"Authorization": "Bearer YOUR_API_KEY"},
    json={
        "model": "volc-asset",
        "Name": "your-custom-name"
    }
)

group_id = response.json()["Id"]
print(f"Group ID: {group_id}")

Expected response:

{
  "Id": "group-20260427160000-xxxxx"
}

Step 2: Create an Asset within the Group

Using the Group ID from Step 1, upload your image asset (e.g., a character reference face).

curl https://www.anyfast.ai/volc/asset/CreateAsset \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "volc-asset",
    "GroupId": "group-20260427160000-xxxxx",
    "Name": "character-reference",
    "AssetType": "Image",
    "URL": "https://example.com/example.png"
  }'

response = requests.post(
    "https://www.anyfast.ai/volc/asset/CreateAsset",
    headers={"Authorization": "Bearer YOUR_API_KEY"},
    json={
        "model": "volc-asset",
        "GroupId": "group-20260427160000-xxxxx",
        "Name": "character-reference",
        "AssetType": "Image",
        "URL": "https://example.com/example.png"
    }
)

asset_id = response.json()["Id"]
print(f"Asset ID: {asset_id}")

Expected response:

{
  "Id": "asset-20260427160000-xxxxx"
}

The URL field accepts three formats:

Public URL: https://example.com/image.jpg

Data URI: data:image/png;base64,iVBOR...

Raw Base64 string (auto-detected, treated as PNG by default)

Base64 / Data URI are automatically uploaded to object storage.

Upload video

Video uploads must specify "model": "volc-asset-video" and "AssetType": "Video".

cURL

curl https://www.anyfast.ai/volc/asset/CreateAsset \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "volc-asset-video",
    "GroupId": "group-20260427160000-xxxxx",
    "Name": "reference-clip",
    "AssetType": "Video",
    "URL": "https://example.com/clip.mp4"
  }'

Upload audio

Audio uploads must specify "model": "volc-asset-audio" and "AssetType": "Audio".

cURL

curl https://www.anyfast.ai/volc/asset/CreateAsset \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "volc-asset-audio",
    "GroupId": "group-20260427160000-xxxxx",
    "Name": "background-music",
    "AssetType": "Audio",
    "URL": "https://example.com/bgm.mp3"
  }'

File upload (multipart)

Upload image

curl https://www.anyfast.ai/volc/asset/CreateAsset \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -F "file=@/path/to/image.jpg" \
  -F "GroupId=group-20260427160000-xxxxx" \
  -F "Name=character-reference"

Upload video

curl https://www.anyfast.ai/volc/asset/CreateAsset \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -F "model=volc-asset-video" \
  -F "file=@/path/to/video.mp4" \
  -F "GroupId=group-20260427160000-xxxxx" \
  -F "Name=reference-clip" \
  -F "AssetType=Video"

Upload audio

curl https://www.anyfast.ai/volc/asset/CreateAsset \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -F "model=volc-asset-audio" \
  -F "file=@/path/to/audio.mp3" \
  -F "GroupId=group-20260427160000-xxxxx" \
  -F "Name=background-music" \
  -F "AssetType=Audio"

Step 3: Generate a Video Using the Asset

Reference the Asset ID from Step 2 to generate a video.

Important: Assets must be passed in strict order: text, image_url, video_url, audio_url. Do not reorder them — doing so may cause errors. When multiple assets are included, do not mix in other asset types.

cURL

curl https://www.anyfast.ai/v1/video/generations \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "seedance",
    "content": [
      {
        "type": "text",
        "text": "The person from @image1 walks in the scene from @video1, with @audio1 as background music"
      },
      {
        "type": "image_url",
        "image_url": {
          "url": "asset://asset-20260427160000-xxxxx"
        },
        "role": "reference_image"
      },
      {
        "type": "video_url",
        "video_url": {
          "url": "asset://asset-20260427160000-video1"
        },
        "role": "reference_video"
      },
      {
        "type": "audio_url",
        "audio_url": {
          "url": "asset://asset-20260427160000-audio1"
        },
        "role": "reference_audio"
      }
    ],
    "resolution": "720p",
    "duration": 5
  }'

The response will return an async task ID (prefixed with asyn).

Step 4: Poll for the Result

Use the task ID to check the generation status.

cURL

curl https://www.anyfast.ai/v1/video/generations/asynxxxx \
  -H "Authorization: Bearer YOUR_API_KEY"

Once complete, the response will contain a pre-signed S3 download link. Please note:

The download link expires after 12 hours.
If the task reaches 100% progress but returns an error, this typically means the output was flagged by the provider’s content moderation (e.g., celebrity likeness or copyrighted IP). In this case, try modifying your prompt or using a different reference image.

Query assets

List asset groups with combined filters

cURL

curl https://www.anyfast.ai/volc/asset/ListAssetGroups \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "volc-asset",
    "Filter": {
      "Name": "codex-seedance-assets-20260723",
      "GroupIds": ["group-20260723040610-zDYKrgRaKt"],
      "Id": "group-20260723040610-zDYKrgRaKt",
      "GroupType": "AIGC"
    },
    "PageNumber": 1,
    "PageSize": 10
  }'

200

{
  "Items": [
    {
      "GroupType": "AIGC",
      "Id": "group-20260723040610-zDYKrgRaKt",
      "Name": "codex-seedance-assets-20260723",
      "Status": "Active"
    }
  ],
  "PageNumber": 1,
  "PageSize": 10,
  "TotalCount": 1
}

List assets in a group

cURL

curl https://www.anyfast.ai/volc/asset/ListAssets \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "volc-asset",
    "Filter": {
      "Name": "character",
      "GroupIds": ["group-20260427160000-xxxxx"],
      "GroupType": "AIGC"
    },
    "PageNumber": 1,
    "PageSize": 10
  }'

Manage assets

Update an asset group

cURL

curl https://www.anyfast.ai/volc/asset/UpdateAssetGroup \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "volc-asset",
    "Id": "group-20260427160000-xxxxx",
    "Name": "updated-project-assets"
  }'

Update an asset

cURL

curl https://www.anyfast.ai/volc/asset/UpdateAsset \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "volc-asset",
    "Id": "asset-20260427161000-xxxxx",
    "Name": "updated-character-reference",
    "GroupId": "group-20260427160000-xxxxx"
  }'

Delete an asset

cURL

curl https://www.anyfast.ai/volc/asset/DeleteAsset \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "volc-asset",
    "Id": "asset-20260427161000-xxxxx"
  }'

Delete an asset group

cURL

curl https://www.anyfast.ai/volc/asset/DeleteAssetGroup \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "volc-asset",
    "Id": "group-20260427160000-xxxxx"
  }'

Billing

Operation	Model	Billed
CreateAssetGroup	`volc-asset`	Free
CreateAsset (image)	`volc-asset`	Free
CreateAsset (video)	`volc-asset-video`	Free
CreateAsset (audio)	`volc-asset-audio`	Free
CreateVisualValidateSession	—	Free
GetVisualValidateResult	—	Free
ListAssetGroups	—	Free
ListAssets	—	Free
GetAsset	—	Free
UpdateAssetGroup	`volc-asset`	Free
UpdateAsset	`volc-asset`	Free
DeleteAsset	`volc-asset`	Free
DeleteAssetGroup	`volc-asset`	Free

Data isolation

When using token-based access, the system automatically prefixes asset group names with [u-{userID}]-[t-{tokenID}], providing user and token-level data isolation. Queries automatically filter to return only data accessible by the current token.

API Reference

Create Asset Group

Create a new asset group.

Create Asset

Upload an asset (image, video, audio) to a group.

Create Liveness Session

Create a mobile liveness face validation session.

Get Liveness Result

Query the LivenessFace group created by a validation session.

List Asset Groups

Query asset groups.

List Assets

Query assets within groups.

Update Asset Group

Update asset group information.

Update Asset

Update asset information.

Delete Asset

Delete an asset.

Delete Asset Group

Delete an asset group.

​Supported Seedance models

​Why use Asset Management?

​Supported formats

​Seedance 2.5 enhanced input capabilities

​Image requirements

​Video requirements

​Audio requirements

​Retention periods

​Rate limits

​Billing models

​Asset review and activation time

​Asynchronous processing and asset URLs

​Workflow

​LivenessFace workflow

​Step 1: Create an Asset Group

​Step 2: Create an Asset within the Group

​Upload video

​Upload audio

​File upload (multipart)

​Step 3: Generate a Video Using the Asset

​Step 4: Poll for the Result

​Query assets

​List asset groups with combined filters

​List assets in a group

​Manage assets

​Update an asset group

​Update an asset

​Delete an asset

​Delete an asset group

​Billing

​Data isolation

​API Reference

Create Asset Group

Create Asset

Create Liveness Session

Get Liveness Result

List Asset Groups

List Assets

Update Asset Group

Update Asset

Delete Asset

Delete Asset Group

Supported Seedance models

Why use Asset Management?

Supported formats

Seedance 2.5 enhanced input capabilities

Image requirements

Video requirements

Audio requirements

Retention periods

Rate limits

Billing models

Asset review and activation time

Asynchronous processing and asset URLs

Workflow

LivenessFace workflow

Step 1: Create an Asset Group

Step 2: Create an Asset within the Group

Upload video

Upload audio

File upload (multipart)

Step 3: Generate a Video Using the Asset

Step 4: Poll for the Result

Query assets

List asset groups with combined filters

List assets in a group

Manage assets

Update an asset group

Update an asset

Delete an asset

Delete an asset group

Billing

Data isolation

API Reference