Stevai API

One model, one endpoint, one key. Send a prompt from your own code and get an answer back — with your region control and transparent, real‑cost pricing built in.

Base URL

Every request goes to:

base url
https://stevai.com

Authentication

Requests are authenticated with an API keysent as a bearer token. You create keys in your account and they're tied to it — usage is billed to you, and never to a shared account. See Authentication for details.

header
Authorization: Bearer sk_live_your_key_here

What's different

  • One model. No model to pick, no version numbers to track. There's exactly one — stevai-default — and it fits every job.
  • Region control. Choose which region processes a request, or leave it global to use the best available.
  • Transparent pricing. Every response reports the real electricity + internet cost of that request — you pay for power, not opaque “tokens.”
  • Session‑safe. The API key authenticates server‑to‑server. Keep it on your backend — never ship it in a browser or mobile app.

Ways to call it

  • Our SDK (recommended). @stevai-ai/sdk for JavaScript/TypeScript and pip install stevai for Python — no third‑party AI library. See SDKs.
  • Drop-in HTTP. The standard chat-completions format — switch an existing app to Stevai by changing only the base URL and key. Supports streaming. See Drop-in API.
  • Simple. A smaller one-prompt/one-answer shape with the full usage + cost breakdown attached. See Chat API.
endpoints
POST https://stevai.com/api/v1/chat/completions   # drop-in
POST https://stevai.com/api/v1/chat               # simple

Next

Head to the Quickstart to create a key and make your first call in under a minute.

NextQuickstart →