imagination AI
Free while in beta
One endpoint.
Every image model.
Call DALL·E, Flux, NanoBanana, Elizie AI and dozens of style models with the same request shape. Pick your model with a single field and we handle routing, retries and fallbacks.
20+
Image models
1
Endpoint to learn
0
Cost while in beta
curl -X POST https://visionary-vision-api.lovable.app/api/public/v1/images \
-H "x-api-key: IMG-RAGE-XXXX-XXXXXX" \
-H "Content-Type: application/json" \
-d '{"prompt":"a neon cyberpunk fox","model":"dall-e-3-xl"}'Why build on Imagination AI
One request shape
Every model takes the same JSON body. Swap a single `model` field to move from photoreal to anime — no rewrites, no new SDKs.
Automatic fallback
If a provider slows down or errors, we retry and reroute to a healthy backend so your app keeps returning images.
Keys in seconds
No signup form, no credit card. Open the Telegram bot, tap one button and your key is ready in the chat.
Hashed key storage
Keys are stored as irreversible SHA-256 hashes. Lost a key? Revoke and regenerate from the bot instantly.
Fair rate limits
Every key gets a per-minute budget with X-RateLimit headers and clear 429 responses — no silent throttling.
Works anywhere
Plain HTTPS + JSON with CORS enabled. Call it from Node, Python, PHP, Flutter, n8n or straight from the browser.
How it works
01
Get your key
Tap 🔑 Create API key in the Telegram bot.
02
Pick a model
Browse the live model list on the docs page or via GET /api/public/v1/models.
03
Send a prompt
POST your prompt with the x-api-key header.
04
Get an image URL
Receive a hosted image URL plus latency and model metadata.
Available models
The list is live — anything enabled below can be called today with the matching model slug.
Questions
Is it really free?
Yes — while Imagination AI is in beta every key is free and unlimited in duration, subject to a per-minute rate limit.
Do keys expire?
No. Keys stay active until you revoke them from the Telegram bot or an administrator suspends them.
Can I use it commercially?
You can integrate the API into your own products. You are responsible for the prompts you send and how you use the output.
What happens if a model is down?
We retry, then fall back to another backend automatically. The response always tells you which model produced the image.
Start generating in under a minute
Grab a free key from the Telegram bot, then send your first prompt straight from the docs playground.