One API.
Best result.
Lowest cost to get there.
Stop choosing between AI models. Describe your task, and our routing engine automatically selects the best model for the job — optimizing for quality and cost so you don't have to.
Quick start
Describe the task. We pick the model.
curl -X POST https://api.getsupermodel.ai/v1/generate \
-H "Authorization: Bearer YOUR_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"task": "product-photo",
"image": "https://your-cdn.com/product.jpg",
"prompt": "Place product on marble countertop, soft studio lighting",
"output": {
"format": "png",
"width": 1024,
"height": 1024
}
}'{
"id": "gen_8f2k9d3j",
"status": "completed",
"model_selected": "flux-pro",
"reason": "Best match for product-photo task at optimal cost",
"result": {
"url": "https://cdn.getsupermodel.ai/gen/8f2k9d3j/output.png",
"latency_ms": 3200,
"quality_score": 0.96
},
"cost": {
"credits_used": 1,
"savings_vs_manual": "42%"
}
}Why Supermodel API
Built for production workloads
Intelligent model selection
Describe your task — our engine evaluates every available model and picks the one that delivers the best result for that specific job.
Optimized for cost
We don't just pick the best model — we pick the best model at the lowest price point. You get top quality without overpaying.
Task-aware routing
Product photos, background removal, upscaling — each task type has different optimal models. Our router knows which one to use and why.
Enterprise-grade reliability
Automatic failover between models. If the top pick is unavailable, your request routes to the next best option. 99.9% uptime SLA.
Global edge delivery
Generated assets are served from a global CDN. Sub-50ms delivery to any region. Optimized formats and compression included.
Developer-first DX
OpenAPI spec, SDKs for Python, Node, and Go. Webhook callbacks, batch processing, and async queues out of the box.
Under the hood
We evaluate. You get the best.
Our routing engine benchmarks every model against your task type. As new models launch, they're automatically evaluated — your results keep getting better without changing a line of code.
Image Generation
Video Generation
API Pricing
Pay as you go
No subscriptions, no tiers. Every generation is routed to the best model for the job — starting at $0.08 per photo and $0.10 per second of video.
Always the best model · Always the best price
- Smart routing to 28+ models
- Best model selected per task automatically
- No markup on model costs
- 4K+ resolution support
- Batch processing & webhooks
- Usage-based — only pay for what you generate
Price varies by model complexity. Our router optimizes cost automatically.
Volume?
Need 10K+ generations/month? Contact us for volume pricing
Start building today
50 free credits to get started. No credit card required. Let our engine find the best model for every task you throw at it.