Image generation and editing

Overview

AGTCloud provides image generation and editing capabilities. After selecting a model, submit text or image input through the OpenAI Images-compatible API and read the generated result.

Supported tasks

Task	Description	Main input
Image generation	Create an image from a text description	Model, prompt
Image editing	Modify an existing image or a selected region	Image, prompt, optional mask
Image variations	Create alternate versions from a source image	Source image

Featured models

Model	Provider	Highlights
Qwen Image	Alibaba Cloud	Supports creative image workflows with multilingual prompts
Seedream	ByteDance	Built for creative visuals, posters, and photorealistic images
Wan Image	Alibaba Cloud	Covers text-to-image, image-to-image, and visual creation workflows

View all models

API access

Base URL

https://api.agtcloud.cn/v1

Generate an image

curl 'https://api.agtcloud.cn/v1/images/generations' \
  -H 'Authorization: Bearer your-api-key' \
  -H 'Content-Type: application/json' \
  --data '{
    "model": "qwen-image",
    "prompt": "A modern library overlooking the sea in soft morning light"
  }'

Edit an image

curl 'https://api.agtcloud.cn/v1/images/edits' \
  -H 'Authorization: Bearer your-api-key' \
  -F 'model=MODEL_ID' \
  -F 'image=@input.png' \
  -F 'mask=@mask.png' \
  -F 'prompt=Replace the background with a beach at sunrise'

Create image variations

curl 'https://api.agtcloud.cn/v1/images/variations' \
  -H 'Authorization: Bearer your-api-key' \
  -F 'model=MODEL_ID' \
  -F 'image=@input.png' \
  -F 'n=2'

Handle image results

Response form	Handling
URL	Download the image and save it to your own storage when needed
Base64	Decode the value into binary image data and save it

The response fields depend on the request parameters and model capabilities. For batch output, process each item in data[].