Docs

Lake AI Service

Video generation

Video Editing Enhanced Edition

The enhanced version of the video editing online service combines input videos and reference images to perform targeted video replacement. It is suitable for scenarios such as advertising material remastering, product replacement, background redrawing, and creative video editing.

Operator introduction

Operator ID: las_video_seedance_replace

Application scenarios

Core capabilities

The current version supports several types of intelligent video editing tasks, including scene replacement (scene_replace), object replacement (object_replace), and person replacement (person_replace).
- Scene replacement (scene_replace): Retains the people, products, actions, and shot structure in the original video, while replacing the background or environmental style.
- Object replacement (object_replace): Replaces the target object in the video with a specified style or product based on a reference image.
- Person replacement (person_replace): Replace the target person in the video with a specified person based on a reference image.
  warning
  - To use the person replacement feature, you must apply to enable the LAS asset managementfunction (allowlist feature, disabled by default) and set seedance_use_asset_api = true when submitting a person replacement task.
  - Required audio, video, or image assets for person replacement:
    - You can manually upload assets to the LAS asset library first, and then provide the asset to the operator by specifying the LAS asset ID when submitting the task.
    - You can also provide assets directly to the operator via HTTPS or TOS paths. In this scenario, LAS will automatically upload the asset to LAS asset management and use the asset ID as input. After the operator task is completed, the automatically uploaded asset will be cleaned up.
Ensures the quality and stability of video editing results.
- Supports supplementing natural language editing constraints via user_prompt, such as retaining actions, removing watermarks, avoiding subtitles, or controlling details of the style.
- Supports automatic evaluation and retry for a limited number of times, helping to achieve more stable replacement results with complex materials.
- Supports returning the final video result and diagnostic file (result.json), making it convenient for the user to retain generated results and process information.

Effect showcase

[Person replacement]

Input video - Example 1

Reference image

Replacement result

Input video - Example 2

Reference image

Replacement result

[Environment and object replacement]

Input video

Scenario 1: Background replacement

Reference image
Output video

Scenario 2: Object replacement

Reference image
Output video

Supported regions

Beijing: cn-beijing
Shanghai: cn-shanghai
Guangzhou: cn-guangzhou

Operator performance

Sub-items	Performance impact description
Maximum RPM	600
Maximum concurrency	10
Processing time	Total processing time depends on the video duration, the number of reference images, the complexity of the replacement, whether retry is enabled, and the network environment. Enabling retry (`enable_retry`) increases the success rate and stability of generation, but significantly increases overall processing time and cost.
Processing result quality	The quality of video content replacement processing results depends on whether a reference image is provided and the content of the reference image. It is recommended to provide reference images as three-view diagrams plus detail close-ups. Especially for scene replacement, keeping the reference image's perspective and distance consistent with the original video can significantly increase the probability of successful replacement.

Input and output requirements

Input requirements

Subdivision	Details
Supported input data modalities	Video Image Audio
Input format: Video	Format: mp4, mov Size: `2-15` seconds. The input size for a single video must not exceed 200 MB. Resolution: 480p, 720p, 1080p Dimensions: Aspect ratio (width/height): [0.4, 2.5] Width and height (px): [300, 6000] Total pixel count: [640x640=409600, 2206×946=2086876], meaning the product of width and height must fall within the range [409600, 2086876]. Other: It is recommended to use video material with a clear subject and a relatively stable camera.
Input format: Image	Format: jpeg, png, webp, bmp, tiff, gif, heic, and heif. Quantity: Supports input of `1-9` reference images. Size: Each image must be less than 30 MB. Resolution: Width and height (px): (300, 6000); aspect ratio (width/height): (0.4, 2.5) Other: It is recommended that the reference images provided include three-view diagrams and detailed close-ups.
Input format: Audio	Format: wav, mp3 Size: Audio duration [2,15] seconds. Each audio file must not exceed 15 MB.
Input path requirements	Provide input data to the operator via request parameters `video_url`, `image_urls`, and `audio_url`. Currently, the following methods are supported: public URL, TOS path, and Base64 encoding (only supported for images and audio). Public URL: A video URL accessible via the public network, with the format `http/https`. TOS path: After uploading the data to be processed to a TOS Bucket under the same primary account and region as the current LAS service, you can provide the TOS path, with the format `tos://bucket_name/demo`. Base64 encoding (only supported for images and audio): Convert the local file to a Base64 encoded string, then submit it to the model. Do not use Base64 encoding for large files. Image input must follow the format: `data:image/<image format>;base64,<Base64 encoding>`. Note that `<image format>` must be in lowercase, for example, `data:image/png;base64,{base64_image}`. Audio input must follow the format: `data:audio/<audio format>;base64,<Base64 encoding>`. Note that `<audio format>` must be in lowercase, for example, `data:audio/wav;base64,{base64_audio}`. LAS asset ID: The ID used for generating real and virtual portraits in videos, following the format: `asset://<asset-id>` . When using the LAS asset ID method, you must first apply to enable the LAS asset managementfunction (allowlist feature, disabled by default), and set `seedance_use_asset_api = true` when submitting the task. When submitting various types of tasks, if the input material involves real-person audio, video, or images, you must first upload the material to the LAS asset library and provide the corresponding material to the model using the LAS asset ID. Otherwise, the task may be blocked. After enabling the feature and uploading materials, you can view the list of asset groups in the LAS Console under asset management > Virtual Portrait Asset. On the asset group details page, you can obtain the specific asset's asset ID (`asset-id`), which can then be concatenated in the above format to be used as input for the operator.

Output requirements

Subdivision	Details
Output data modality	Video Text (JSON file)
Output format: Video	Size: The output video duration is within the range of `4-15` seconds. Aspect ratio: Original video ratio (default), 16:9, 9:16, 1:1, 4:3, 3:4, 21:9, 9:21. Resolution: Supports setting to 480p, 720p, 1080p. If not set, the output will be processed automatically according to the input video. If a unified output resolution is required, it is recommended to set the output resolution value via the `seedance_resolution` parameter. If the output video resolution is not explicitly specified, it will match the input video resolution.
Output path: TOS	Specify the storage path for the operator's output results via the request parameter `output_tos_path`. Currently, setting to a TOS path is supported. TOS path: A TOS Bucket directory with write permissions under the same primary account and region as the LAS service.

Subdivision

Details

Output data modality

Video
Text (JSON file)

Output format: Video

Size: The output video duration is within the range of 4-15 seconds.
Aspect ratio: Original video ratio (default), 16:9, 9:16, 1:1, 4:3, 3:4, 21:9, 9:21.
Resolution: Supports setting to 480p, 720p, 1080p. If not set, the output will be processed automatically according to the input video.
- If a unified output resolution is required, it is recommended to set the output resolution value via the seedance_resolution parameter.
- If the output video resolution is not explicitly specified, it will match the input video resolution.

Output path: TOS

Specify the storage path for the operator's output results via the request parameter output_tos_path. Currently, setting to a TOS path is supported.

TOS path: A TOS Bucket directory with write permissions under the same primary account and region as the LAS service.

Billing instructions

Billing standards

Sub-items	Billing standards description
Billing item	Duration of input/output video.
Billing type	Pay-as-you-go billing, unit: CNY/second, billed hourly according to actual usage.
Unit price	Depends on the version of the service you select (basic version or enhanced version).

warning

Total cost paid: In addition to the unit price, the total cost also depends on the duration and resolution of the input and output videos.
Only tasks that successfully generate videos will incur costs. No charges will be applied for scenarios where video generation fails due to review or other operational reasons.

Billing details
Billing formula: Total cost = unit price * billing volume = unit price * (video duration * duration conversion coefficient)

Version breakdown	Unit price	Scenario breakdown	Billing item: video duration	Duration conversion coefficient: is determined by video resolution
Basic version	1.6 CNY/second	No input video	Output video duration	480P：0.465 720P：1
Basic version	1.6 CNY/second	With input video	(Input video duration + output video duration)/2	480P：0.554 720P：1.220
Enhanced version	2 CNY/second	No input video	Output video duration	480P：0.465 720P：1 1080P：2.504
Enhanced version	2 CNY/second	With input video	(Input video duration + output video duration)/2	480P：0.580 720P：1.220 1080P：3.040

Billing example
- Example scenario: Using the enhanced version of video generation, the input includes a reference video (duration: 5 seconds), the output video duration is 7 seconds, and the resolution is 1080p.
- Cost details:
  Total cost = unit price * billing volume = 2 CNY/second * (5 seconds + 7 seconds)/2 * 3.040 = 36.48 CNY
Billing unit
The billing unit displayed in the billing details for costs incurred by invoking this operator is "Video editing enhanced version". Please pay attention to this difference when reviewing your bill.

Caution and prerequisites

Details	Caution and prerequisites
Costs	Before calling an operator, you need to understand the model invocation costs associated with using the operator. For details, see Large model invocation billing.
Authentication (API Key)	Before calling an operator, you need to generate an API Key for operator invocation. It is recommended to configure the API Key as an environment variable to ensure safer operator calls. For details, see Obtain and configure API Key.
BaseURL	Before calling an operator, you need to determine the BaseURL for operator invocation based on the region where your current LAS service is deployed. This is used to configure the path parameter values for operator calls. For details, see Obtain the Base URL. The Examples below are for reference only; when making actual calls, replace the path values with those corresponding to your region.

Rest API call

API usage instructions

The operator's API is an asynchronous task interface. You need to submit a task through the submit interface, then use the Poll interface to obtain the task status and retrieve the operator's processing results. Additionally, the operator supports asynchronous callback capability (Callback). You can set the callback parameter callback.url when initiating the operator task. LAS will send an HTTP POST notification to this URL. For guidance on using asynchronous callbacks, see Asynchronous callback API.

Submit

Interface description

You can submit asynchronous tasks by invoking the video editing enhanced version operator. This supports scene replacement, object replacement, and other video replacement editing operations based on input video and reference images or audio, and uploads the results to the specified TOS directory.

Request parameters

参数	类型	Required	示例值	描述
operator_id	string	Yes	las_video_seedance_replace	Operator ID
operator_version	string	Yes	v1	Operator version
data	VideoSeedanceReplaceReqParams	Yes		Video replacement request parameters
callback	Callback	否		Task status callback configuration. After configuring callback.url, a POST request will be sent to this address when the task status changes. If not configured, you can query the task status and result using Poll.

Response parameters

参数	类型	Example value	Description
metadata	metadata		Request metadata

Request example

# Please set OUTPUT_TOS_PATH to a tos path writable by this account
export OUTPUT_TOS_PATH="tos://path/to/seedance_replace_results"

curl --location "https://operator.las.ap-southeast-1.volces.com/api/v1/submit" \
--header "Content-Type: application/json" \
--header "Authorization: Bearer $LAS_API_KEY" \
--data '{
    "operator_id": "las_video_seedance_replace",
    "operator_version": "v1",
    "data": {
        "video_url": "https://${LAS_TOS_BUCKET}.${TOS_ENDPOINT}/operator_cards_serving/public/${LAS_ENVIRONMENT}/las_video_seedance_replace/v1/sample_input.mp4",
        "image_urls": [
            "https://${LAS_TOS_BUCKET}.${TOS_ENDPOINT}/operator_cards_serving/public/${LAS_ENVIRONMENT}/las_video_seedance_replace/v1/sample_reference.jpg"
        ],
        "output_tos_path": "$OUTPUT_TOS_PATH",
        "user_prompt": "请帮我把@视频1的背景替换成@图片1的风格，保持人物动作和商品一致性，除背景外不做任何更改，不要字幕。",
        "task_type": "scene_replace",
        "enable_retry": true,
        "max_retry_times": 2
    }
}'

Response example

{
  "metadata": {
    "task_id": "task-20260427143000-abc123",
    "task_status": "PENDING",
    "business_code": "0",
    "error_msg": ""
  }
}

Error code

HttpStatusCode	Error code	Error message	Description
400	Url.Invalid	The url is invalid.	The input video, reference image, or audio link is invalid.
400	Video.Invalid	The video file is invalid.	The video file is invalid or cannot be parsed.
400	Video.DownloadFailed	Video download failed.	Video download failed.
400	Video.FormatUnsupported	Video format not supported.	Video format not supported.
400	Video.DurationTooShort	Video duration is too short.	Video duration is below the minimum limit.
400	Video.DurationExceeded	Video duration exceeds maximum limit.	The video duration exceeds the maximum limit.
400	Video.FileTooLarge	Video file size exceeds maximum limit.	The video file is too large.
400	Parameter.Invalid	The parameter is invalid.	The parameter is invalid.
400	Video.Timeout	Video processing timeout.	Video processing timed out.
400	Video.UploadFailed	Video upload failed.	Failed to upload the resulting video.
400	Tos.AccessFailed	TOS access failed.	TOS access failed.
401	Authorization.Missing	Missing Authorization.	Authorization is missing.
401	ApiKey.InValid	The api key is invalid.	The API is invalid.
500	Internal.ServerError	Internal server error.	Internal server error, such as failure in generation, evaluation, or result assembly.

Poll

Interface description

Query the execution status and result of the video replacement task.

Request parameters

Name	Type	Required	Example value	Description
operator_id	string	Yes	las_video_seedance_replace	Operator ID
operator_version	string	Yes	v1	Operator version
task_id	string	Yes	task-xxx	Task ID

Response parameters

参数	类型	示例值	描述
metadata	metadata		Request metadata
data	VideoSeedanceReplaceResponse		Returned data

Request example

curl --location "https://operator.las.ap-southeast-1.volces.com/api/v1/poll" \
--header "Content-Type: application/json" \
--header "Authorization: Bearer $LAS_API_KEY" \
--data '{
    "operator_id": "las_video_seedance_replace",
    "operator_version": "v1",
    "task_id": "task-20260427143000-abc123"
}'

Response example

{
    "metadata": {
        "task_id": "task-20260427143000-abc123",
        "task_status": "COMPLETED",
        "submit_time": "2026-07-21T16:04:06+08:00",
        "end_time": "2026-07-21T16:04:06+08:00",
        "business_code": "0",
        "error_msg": ""
    },
    "data": {
        "output_video_tos_url": "tos://path/to/seedance_replace_results/run_id/generated/attempt_01.mp4",
        "output_video_url": "https://path/to/seedance_replace_results/run_id/generated/attempt_01.mp4",
        "selected_attempt_no": 1,
        "attempt_count": 1,
        "output_result_json_tos_url": "tos://path/to/seedance_replace_results/run_id/result.json",
        "output_result_json_url": "https://path/to/seedance_replace_results/run_id/result.json"
    }
}

Error code

HttpStatusCode	Error code	Error message	Description
400	Url.Invalid	The url is invalid.	The input video, reference image, or audio link is invalid.
400	Video.Invalid	The video file is invalid.	The video file is invalid or cannot be parsed.
400	Video.DownloadFailed	Video download failed.	Video download failed.
400	Video.FormatUnsupported	Video format not supported.	Video format not supported.
400	Video.DurationTooShort	Video duration is too short.	Video duration is below the minimum limit.
400	Video.DurationExceeded	Video duration exceeds maximum limit.	The video duration exceeds the maximum limit.
400	Video.FileTooLarge	Video file size exceeds maximum limit.	The video file is too large.
400	Parameter.Invalid	The parameter is invalid.	The parameter is invalid.
400	Video.Timeout	Video processing timeout.	Video processing timed out.
400	Video.UploadFailed	Video upload failed.	Failed to upload the resulting video.
400	Tos.AccessFailed	TOS access failed.	TOS access failed.
401	Authorization.Missing	Missing Authorization.	Authorization is missing.
401	ApiKey.InValid	The api key is invalid.	The API is invalid.
500	Internal.ServerError	Internal server error.	Internal service error.

Last updated: 2026.08.06 09:49:44

Lake AI Service

Application scenarios #

Core capabilities #

Effect showcase #

Supported regions #

Operator performance #

Input requirements #

Output requirements #

API usage instructions #

Submit #

Interface description #

Request parameters #

Response parameters #

Request example #

Response example #

Error code #

Poll #

Interface description #

Request parameters #

Response parameters #

Request example #

Response example #

Error code #

Application scenarios

Core capabilities

Effect showcase

Supported regions

Operator performance

Input requirements

Output requirements

API usage instructions

Submit

Interface description

Request parameters

Response parameters

Request example

Response example

Error code

Poll

Interface description

Request parameters

Response parameters

Request example

Response example

Error code