tencent cloud

LLM Service TokenHub

ドキュメントLLM Service TokenHubAPI Integration GuideImage Generation ModelsWAND Image Generation API Guide

WAND Image Generation API Guide

ダウンロード
フォーカスモード
フォントサイズ
最終更新日: 2026-09-16 11:24:06
AI翻訳

Overview

Tencent Cloud WAND (Tencent Cloud Audio/Video AI) is an AI capability brand built by Tencent Cloud for audio/video scenarios. Audio/video scenarios are complex and diverse, making it difficult for general-purpose AI to deliver results directly. Leveraging Tencent Cloud's years of audio/video technology expertise and extensive business practices, WAND deeply integrates multimodal large models with audio/video engineering experience to provide AI atomic capabilities validated in real-world scenarios. WAND-Vega-Image is a cost-effective multimodal image generation model that offers three tiers of capabilities: Lite, Flash, and Pro. Whether for high-volume content production that prioritizes efficiency or for professional creation that demands higher image quality, you can choose the appropriate tier as needed to achieve stable and efficient image generation results at a controlled cost.

Prerequisites

You have registered a Tencent Cloud account and activated the TokenHub service.
You have obtained the API Key from the TokenHub console.
Note:
In all examples below, YOUR_API_KEY must be replaced with your own API Key. The authentication method is the request header Authorization: Bearer YOUR_API_KEY.

Call Process

Image generation is a time-consuming task, so the API uses an asynchronous call model, which is divided into two steps:
1. Submit a task: Call the image generation API, which returns a task_id upon success.
2. Poll for results: Call the Querying the Task Result API with the task_id until status = completed, and then obtain the image URL from the result.

Model List

Note:
The WAND-Vega image generation model offers three tiers of capability models: Lite, Flash, and Pro, supporting text-to-image and reference-to-image generation. Input images are free for the Lite and Flash tiers. For the Pro tier, the first 3 input images are free, and input image fees are charged per image starting from the 4th image.
Model Name
Model (API Parameter)
Supported Capability
Prompt Limit
Resolution
Aspect Ratio
WAND-Vega-Image-1.0-Lite
wand-vega-image-lite
Text-to-image/reference-based image generation
2000 characters
1K / 2K / 4K
Supports common sizes such as 1:1, 2:3, 3:2, 3:4, 4:3, 9:16, 16:9, and 21:9. For details, see List of Preset Sizes.
WAND-Vega-Image-1.0-Flash
wand-vega-image-flash
Text-to-image/reference-based image generation
2000 characters
1K / 2K / 4K
WAND-Vega-Image-1.0-Pro
wand-vega-image-pro
Text-to-image/reference-based image generation
2000 characters
1K / 2K / 4K

Image Generation

1. API Description

The WAND-Vega image generation API supports text-to-image and reference-to-image generation: it generates images from an input reference image + text prompts, and performs text-to-image generation when no image is provided.
API: POST https://tokenhub-intl.tencentcloudmaas.com/v1/wand/vega-images/generations

2. Input Parameters

Parameter Name
Required
Type
Description
model
Yes
string
Model ID. Values:
wand-vega-image-lite
wand-vega-image-flash
wand-vega-image-pro
prompt
Yes
string
Text prompt, with a length of no more than 2000 characters. If images is not passed, image generation is performed based on this text.
input
No
array[object]
Array of multimodal input content, used for image-to-image / reference image scenarios.
input.content
No
array[object]
Array of input content segments, concatenated in order to form a multimodal prompt.
input.content.type
No
string
Content type. Values: input_image (reference image).
input.content.image_url
No
string
Reference image URL. lite: 0 to 3 images, flash/pro: 0 to 6 images. Format: png/jpeg. Size: no more than 10 MB.
size
No
string
Output image size, in the format ${width}x${height}.

3. Sample Request

Text-to-image
curl -X POST 'https://tokenhub-intl.tencentcloudmaas.com/v1/wand/vega-images/generations' \\
  -H 'Authorization: Bearer YOUR_API_KEY' \\
  -H 'Content-Type: application/json' \\
  -d '{
  "model": "wand-vega-image-lite",
  "prompt": "A detailed painting of a butterfly",
  "size": "1024x1024"
}'
Reference-to-image
curl -X POST 'https://tokenhub-intl.tencentcloudmaas.com/v1/wand/vega-images/generations' \\
  -H 'Authorization: Bearer YOUR_API_KEY' \\
  -H 'Content-Type: application/json' \\
  -d '{
  "model": "wand-vega-image-lite",
  "prompt": "A seamless composition of all reference images",
  "input": [
    {
      "content": [
        {
          "type": "input_image",
          "image_url": "https://example.com/reference.jpg"
        }
      ]
    }
  ],
  "size": "1024x1024"
}'

4. Output Parameters

Parameter Name
Type
Description
task_id
string
Task ID generated by WAND-Vega image generation, used for subsequent task queries and callback matching.
request_id
string
Unique request identifier used for troubleshooting.

5. Sample Response

{
  "task_id":"251380669-VegaImage-f91e7435e1e34dfa9898a451946bb3d3",
  "request_id":"5b7c57a3-e5e4-4df0-958b-e72d35582be1"
}

Querying the Task Result

1. API Description

Query the status and results of an image generation task. After submitting a task and receiving a task_id, poll this API to obtain the results.
API: GET https://tokenhub-intl.tencentcloudmaas.com/v1/wand/vega-images/tasks/<task_id>
Note:
The {task_id} in the path is the task_id returned when the task is submitted (in the example, YOUR_TASK_ID is used as a placeholder). Image generation takes approximately several seconds to tens of seconds, so it is recommended to poll every 3 to 5 seconds.

2. Input Parameters

Parameter Name
Required
Type
Description
task_id
Yes
string
Task ID (path parameter), which is the task_id returned when the task is submitted.

3. Sample Request

curl -X GET 'https://tokenhub-intl.tencentcloudmaas.com/v1/wand/vega-images/tasks/<task_id>' \\
  -H 'Authorization: Bearer YOUR_API_KEY' \\
  -H 'Content-Type: application/json'

4. Output Parameters

Field
Type
Description
status
string
Processing status: queued/in_progress/completed/failed/cancelled/incomplete.
queued: The task has been accepted and is queued for execution.
in_progress: The task is being processed.
completed: The task has been completed successfully. You can read data[].url to obtain the generated image.
failed: The task failed to execute.
cancelled: The task has been cancelled.
incomplete: The task was interrupted/not completed (for example, when the token limit is reached).
request_id
string
Unique request identifier used for troubleshooting.
data
array[object]
List of generated results.
data[].url
string
Download URL of the generated image. It is a temporary URL with a signature. Download and store it promptly.
usage
object
Usage consumption.
usage.total_tokens
integer
Number of tokens consumed by this task, used for billing/reconciliation.
created_at
string
Task creation time.
finished_at
string
Task completion time.

5. Sample Response

{
  "task_id": "251380669-VegaImage-68a067d14b3c43c0961ce83373709f44",
  "status": "completed",
  "request_id": "92aa30e9-7549-4319-a0eb-467302d446f5",
  "data": [
    {
      "url": "https://aigc-image.cos.myqcloud.com/xxx/result.png"
    }
  ],
  "usage": {
    "total_tokens": 16200
  },
  "created_at": 1787560807,
  "finished_at": 1787560820
}

Appendix

Querying the Task Result

Aspect Ratio
1K Resolution
2K Resolution
4K Resolution
1:1
1024×1024,1080×1080
2048×2048,2160×2160
4096×4096
2:3
1024×1536,1080×1620
1440×2160
3328×4992
3:2
1536×1024,1620×1080
2160×1440
4992×3328
3:4
1080×1440
1440×1920,2160×2880
3584×4800
4:3
1440×1080
1920×1440,2880×2160
4800×3584
9:16
720×1280,1080×1920
1440×2560
2160×3840
16:9
1280×720,1920×1080
2560×1440
3840×2160
21:9
1512×648
2520×1080
5040×2160

Error Code


前の記事へ: Image Generation Model Invocation Overview次の記事へ: Hy Image Generation API Guide

ヘルプとサポート

この記事はお役に立ちましたか？

フィードバック

キーワードで検索
    Enter keywords to search documentation