Inference, without the overhead

One key. Every model. Clarity on every token.

Ship inference with a single API surface, transparent per-token pricing, and an observability layer built for product teams.

Drop-in compatible

Point your existing SDKs at a new base URL.

One catalog

Discover model limits, capabilities, and pricing.

Server-side auth

Scoped API keys that stay out of the browser.

POST /v1/chat/completions
Authorization: Bearer sk_live_xxxxxxxx
200 OK · usage accounted per token

Console access

Sign in to your account

Manage your models, usage, and API keys in one place.

New to Inference Cloud? Create an account

Looking for API documentation? Read the docs