Skip to content

Vega Video 1.5 Lite โ€‹

Updated

โ† Back to the Vega Video overview for version comparison and selection.

Endpoint: POST https://vod.intl.tencentcloudapi.comAction: CreateAigcVideoTaskInternational API documentation: CreateAigcVideoTask

Vega Video 1.5 Lite (1.5-lite) is the new Vega Lite release, balancing generation quality, speed and cost. It supports text-to-video, first-frame / last-frame / first-and-last-frame generation, and multimodal reference generation from mixed image, video and audio inputs. It outputs up to 4K at 2โ€“30 seconds per clip, also accepts -1 for model-selected duration, and can generate audio with the video.

Basic Information โ€‹

ItemValue
ModelNamewand-vega-video
ModelVersion1.5-lite
Resolution480P / 720P / 1080P / 2K / 4K
Aspect ratio21:9 / 16:9 / 4:3 / 1:1 / 3:4 / 9:16 / adaptive
Duration2โ€“30 seconds, or -1 for model-selected duration
Output soundSupported (AudioGeneration: Enabled)

Input Specifications โ€‹

Input typeCount and durationPer-file size
First / last frame image1 eachUp to 20MB
Reference imagesUp to 10Up to 20MB each
Reference videosUp to 5, no more than 15 seconds in totalUp to 100MB each
Reference audio clipsUp to 5, no more than 15 seconds in totalUp to 15MB each

Reference inputs are passed through FileInfos, with Usage set to FirstFrame, LastFrame or Reference. Submit audio reference together with image or video material.

Request Parameters โ€‹

ParameterRequiredTypeDescription
SubAppIdRequiredIntegerVOD application ID
ModelNameRequiredStringFixed value wand-vega-video
ModelVersionRequiredStringFixed value 1.5-lite
PromptConditionalStringGeneration or editing instruction; required when FileInfos is empty
FileInfos.NOptionalArrayImage, video or audio reference material; Usage is FirstFrame / LastFrame / Reference
OutputConfig.ResolutionOptionalString480P / 720P / 1080P / 2K / 4K
OutputConfig.DurationOptionalFloat2โ€“30, or -1 for model-selected duration
OutputConfig.AspectRatioOptionalString21:9 / 16:9 / 4:3 / 1:1 / 3:4 / 9:16 / adaptive
OutputConfig.StorageModeOptionalStringTemporary / Permanent
OutputConfig.MediaNameOptionalStringMedia name shown when Permanent
OutputConfig.AudioGenerationOptionalStringEnabled / Disabled
ExtInfoOptionalStringJSON-encoded string for passing WandVegaParameters

Request Examples โ€‹

Text-to-Video โ€‹

json
{
  "SubAppId": 123456789,
  "ModelName": "wand-vega-video",
  "ModelVersion": "1.5-lite",
  "Prompt": "A woman walks through a rain-soaked neon street, cinematic camera movement and natural motion.",
  "OutputConfig": {
    "Resolution": "1080P",
    "Duration": 10,
    "AspectRatio": "16:9",
    "AudioGeneration": "Enabled",
    "StorageMode": "Temporary"
  }
}

First and Last Frame Generation โ€‹

json
{
  "SubAppId": 123456789,
  "ModelName": "wand-vega-video",
  "ModelVersion": "1.5-lite",
  "Prompt": "Create a smooth, cinematic transition between the two scenes.",
  "FileInfos": [
    {
      "Type": "Url",
      "Category": "Image",
      "Usage": "FirstFrame",
      "Url": "https://<cdn>/first.png"
    },
    {
      "Type": "Url",
      "Category": "Image",
      "Usage": "LastFrame",
      "Url": "https://<cdn>/last.png"
    }
  ],
  "OutputConfig": {
    "Resolution": "1080P",
    "Duration": 8,
    "AspectRatio": "16:9",
    "StorageMode": "Temporary"
  }
}

Multimodal Reference Generation โ€‹

json
{
  "SubAppId": 123456789,
  "ModelName": "wand-vega-video",
  "ModelVersion": "1.5-lite",
  "Prompt": "Keep the character identity and follow the reference video's camera rhythm.",
  "FileInfos": [
    {
      "Type": "Url",
      "Category": "Image",
      "Usage": "Reference",
      "Url": "https://<cdn>/character.png"
    },
    {
      "Type": "Url",
      "Category": "Video",
      "Usage": "Reference",
      "Url": "https://<cdn>/camera-reference.mp4"
    },
    {
      "Type": "Url",
      "Category": "Audio",
      "Usage": "Reference",
      "Url": "https://<cdn>/music-reference.mp3"
    }
  ],
  "OutputConfig": {
    "Resolution": "2K",
    "Duration": -1,
    "AspectRatio": "adaptive",
    "AudioGeneration": "Enabled",
    "StorageMode": "Temporary"
  }
}

Responses and Result Query โ€‹

The generation request returns a TaskId. Query the task with DescribeTaskDetail. Generation succeeds when AigcVideoTask.Status=FINISH and ErrCode=0; the video URL is returned in AigcVideoTask.Output.FileInfos.FileUrl.

json
{
  "SubAppId": 123456789,
  "TaskId": "<task-id>"
}

Special Settings โ€‹

ExtInfo (WandVegaParameters) โ€‹

Lite accepts model-specific options through the WandVegaParameters key in ExtInfo. Serialize the inner parameters object as a JSON string, place it inside ExtInfo, and serialize the outer object again.

ParameterTypeValue / behavior
EnhancePromptModeStringauto / enable / disable. Omission or auto is treated as enable
ReturnVegaUsageBooleanWhether to return Vega usage in the result; default false
ModerationModeStringModeration_L1 / Moderation_L2

Difference from 1.0 Lite

1.5-lite does not support GenerateMode or VideoQualityMode. Those two parameters are available only on 1.0 Lite.

python
import json

ext_info = json.dumps({
    "WandVegaParameters": json.dumps({
        "EnhancePromptMode": "enable",
        "ReturnVegaUsage": True,
        "ModerationMode": "Moderation_L1",
    })
})

Permanent Storage โ€‹

Set OutputConfig.StorageMode to Permanent to keep the output as a VOD media asset. You can also specify MediaName.

Built with VitePress ยท Deployed on Tencent EdgeOne Pages