tencent cloud

Media Processing Service

Video Fission

Download
Focus Mode
Font Size
Last updated: 2026-09-11 16:15:08
AI-Translated

Feature Description

Video Remixing is a set of product short video generation capabilities. Upload a product image to generate a corresponding short video based on the selected video type (UGC seeding, product voiceover, product showcase, unboxing sharing, reaction showcase). You can also adapt the voiceover and on-screen presenters based on the target market and language to quickly produce product videos that meet advertising requirements.

Prerequisites

Before using this feature, you need to complete the following preliminary operations:
Register or log in to a Tencent Cloud account, activate MPS, and complete the service role authorization.
If you use a Tencent Cloud sub-account, ensure the account has sufficient permissions to use MPS.
For detailed guidance, see Getting Started. For account authorization issues, see the Account Authorization document.

Billing Overview

A fee is charged for "Video Remixing". For detailed pricing, see the Billing Description document.

Creating a Video Fission Task

Call the CreateAiFissionTask API to initiate a video remixing task.
Generation Parameters (TaskInfo)
Parameter
Type
Required
Description
TaskInfo.Duration
Integer
No
Video output duration in seconds. Valid values: 4 to 15. Default value: 15.
TaskInfo.ModelTier
String
No
Model tier:
flagship: Premium edition.
standard: Standard edition.
TaskInfo.Ratio
String
No
Video frame aspect ratio: 9:16 / 16:9 / 1:1 / 3:4 / 4:3.
TaskInfo.Resolution
String
No
Output resolution: 720p / 1080p / 2k / 4k.
TaskInfo.SplitCount
Integer
No
Split quantity. Valid value: 1.
Content and Style Parameters (TaskInfo)
Parameter
Type
Required
Description
TaskInfo.VideoType
String
No
Video type: ugc (UGC recommendations) / talk (product voiceover) / display (product showcase, product-only, no voiceover) / unboxing (unboxing sharing) / reaction (reaction showcase).
TaskInfo.Language
String
No
Voiceover/subtitle language: english (English) / chinese (Chinese) / japanese (Japanese) / korean (Korean) / spanish (Spanish) / portuguese (Portuguese) / music_only (pure music without voiceover).
TaskInfo.Market
String
No
Target market: north_america (North America) / europe (Europe) / china (China) / japan (Japan) / korea (South Korea) / southeast_asia (Southeast Asia) / brazil (Brazil) / global (Global) / other (Other).
Affects the default on-camera model ethnicity and localization style. When CustomModel is not specified, the system automatically determines the model ethnicity based on the market.
Model Persona Parameters (TaskInfo.CustomModel)
Parameter
Type
Required
Description
CustomModel.Gender
String
No
Gender: male/female/any.
CustomModel.Age
String
No
Age range: teen/youth/young_adult/middle_aged/mature.
CustomModel.Appearance
String
No
Appearance: caucasian/asian/latino/african/middle_eastern.
CustomModel.BodyType
String
No
Body type: slim/standard/athletic/chubby.
Initiation example:
{
"Input": {
"ImageUrls": [
"https://cg-1234567.cos.ap-guangzhou.myqcloud.com/test/product.jpg" // Replace it with your reference product image.
],
"Text": "Generate a video introducing a facial mask" // Custom prompt that specifies additional requirements for the generated video.
},
"CosInfo": { // Replace it with your output storage path.
"CosBucketRegion": "ap-guangzhou",
"CosBucketName": "cg-1234567",
"CosBucketPath": "/test/output/"
},
"TaskInfo": {
"Duration": 15, // Duration of the generated video in seconds. Valid values range from 4 to 15.
"ModelTier": "flagship", // Model parameter
"Ratio": "9:16", // Aspect ratio of the generated video.
"Resolution": "1080p", // Resolution of the generated video.
"Market": "china", // Target market
"Language": "chinese", // Target language of the generated video.
"VideoType": "ugc",
"SplitCount": 1,
"CustomModel": { // Specifies the model image for the generated video.
"Gender": "female", // Model gender
"Age": "young_adult", // Model age
"Appearance": "asian", // Model ethnicity
"BodyType": "slim" // Model body type
}
}
}

Querying Tasks

After you initiate a viral video replication task using CreateAiFissionTask, a task ID (TaskId) is returned, for example: 24000191-ViralClone-92e56dab-2a5b-4537-b49f-xxxxxxxxxd4903f. To obtain the task result, call the DescribeAigcTaskStatus API and provide the task ID.
Successful response example:
{
"Response": {
"CreateTime": "2026-08-27 14:52:47",
"FinishedTime": "2026-08-27 15:03:30",
"OutputUrl": "https://test-1234567.cos.ap-guangzhou.myqcloud.com/aifission/output/2600011633-AigcScenario-b44a6c75c466458c54c82a52f6f5650c-202608271503-0.mp4",
"RequestBody": "{\\"Input\\":{\\"ImageUrls\\":[\\"https://test-1234567.cos.ap-guangzhou.myqcloud.com/aifission/product.webp\\"],\\"Text\\":\\"\\"},\\"TaskInfo\\":{\\"Duration\\":15,\\"ModelTier\\":\\"flagship\\",\\"Ratio\\":\\"9:16\\",\\"Resolution\\":\\"1080p\\",\\"Market\\":\\"china\\",\\"Language\\":\\"chinese\\",\\"VideoType\\":\\"ugc\\",\\"SplitCount\\":1},\\"CosInfo\\":{\\"CosBucketRegion\\":\\"ap-guangzhou\\",\\"CosBucketName\\":\\"mpstest-celiaxqli-1308104797\\",\\"CosBucketPath\\":\\"/aifission/output\\"},\\"Action\\":\\"CreateAiFissionTask\\",\\"RequestId\\":\\"2701f527-f80f-4c7d-be0b-7d179f02a493\\",\\"Timestamp\\":\\"1787813567\\",\\"AppId\\":1308104797,\\"RequestSource\\":\\"API\\",\\"RequestClient\\":\\"\\",\\"ApiModule\\":\\"mps\\",\\"Uin\\":\\"100022076939\\",\\"ClientIp\\":\\"81.71.133.172\\",\\"SubAccountUin\\":\\"100041649512\\",\\"Region\\":\\"\\"}",
"RequestId": "816ceeb2-de7f-4bab-8a95-09053d3e8ac7",
"ScheduledTime": "2026-08-27 14:52:48",
"TaskId": "2600011633-AigcScenario-b44a6c75c466458c54c82a52f6f5650c",
"TaskStatus": "FINISHED",
"TaskType": "FissionVideo"
}
}


Help and Support

Was this page helpful?

Help us improve! Rate your documentation experience in 5 mins.

Feedback