ImageGPT API
Unified REST API for generating text, images, video, and audio from 40+ AI models through a single endpoint with credit-based pricing.
Base URL
All requests are made to:
https://api.imagegpt.online
What you can build
| Playground | Endpoint prefix | What it does |
|---|---|---|
| Text | /generate/text | Chat completions from 20+ LLMs (OpenAI, Google, Anthropic, DeepSeek, and more). |
| Image | /generate/text-image, /generate/image-image | Text-to-image and image-to-image generation with 15+ models. |
| Video | /generate/text-video, /generate/image-video | Asynchronous text-to-video and image-to-video with 7+ models. |
| Audio | /generate/text-speech | Text-to-speech from Qwen3, ElevenLabs, and OpenAI. |
How it works
- Get an API key and send it in the
x-api-keyheader on every request. See Authentication. - Pick a model from the model list on each playground page.
- Call the endpoint with a JSON body describing what you want.
- Pay in credits — each model deducts a credit amount per request, per token, per image, or per second depending on the playground.
Credit-based pricing
Every playground deducts credits based on its own unit:
- Text — credits per 1,000 output tokens (input tokens are free).
- Image — credits per generated image.
- Video — credits per second of footage (varies by model and resolution).
- Audio — credits per request.
Each playground page lists the exact credit cost for every available model.