Get a key

The catalog

Every model. One key.

Pin a model by name and get exactly that model, or pass a tier and let the router choose.

  • 1000+Models
  • 50+Labs
  • 4Modalities
  • 1API key

Direct access

Pinned by name, never substituted.

Quality is Corent's score out of ten. Pin a name and you get that model or nothing: no substitution, flat cost-plus, billed only on success.

1000+ models

ImagePOST /v1/images/generate

LabPin by nameQuality

VideoPOST /v1/videos/generate

LabPin by nameQuality

VoicePOST /v1/audio/speech

LabPin by nameQuality

LanguagePOST /v1/chat/completions

LabPin by nameQuality

Two ways to call

Name a tier, or name a model.

The router picks

POST /v1/images/generate{  "prompt": "a lighthouse at dusk",  "tier": "max_pro"}

Five tiers, air through max_pro. A tier is a live menu, not a label on a model, so the receipt names the tier that served you.

You pick

POST /v1/images/generate{  "prompt": "a lighthouse at dusk",  "model": "corent-flux-schnell"}

Direct access runs that model and no other. Video still scales with duration and resolution, snapped to the model's own menu.

On the OpenAI-compatible chat endpoint the tier rides in the model field instead: corent/text-premium.

What each tier costs, and how to call it.

PricingDocs