Image/Orchestrated
The right image
model, every time.
Describe what you want in plain language. Corent reads the prompt, routes it to the best image model for your goal, and returns one clean result. No model names, no parameters, no dead ends.
One prompt.
You never choose an image model again. Corent profiles the prompt, its text, faces, composition and complexity, then routes to the model that wins for your optimization goal at that moment, then reroutes automatically if a provider slips.
Optimize for
Say the goal. We route the rest.
Fast
corent-image-fast
Drafts and iterations in seconds. The cheapest route that still holds the composition.
Balanced
corent-image-balanced
The default. Corent weighs quality against price and picks the model that wins on both.
Premium
corent-image-premium
Maximum fidelity: clean text, faithful faces, print-ready detail, routed to the strongest model live.
Receipts
Real generations. Real numbers.
One endpoint
A sentence in.
An image out.
No SDKs per provider, no model IDs to track. Send a prompt and how to optimize it; Corent returns the image and the receipt.
Read the docs// one call, plain language { "prompt": "a matte black espresso machine on carrara marble", "optimize": "quality" }