OneAPIKey
Documentation

Docs

Start here. Full SDK reference, streaming, tool-calling, and SSO integration follow.

Quickstart
Create an account, generate a key, and make your first request in under 90 seconds.
OpenAI compatibility
Point your existing OpenAI SDK at OneAPIKey's base URL. Everything just works.
Model IDs
Use `provider/model` (e.g. `anthropic/claude-opus-4-7`) or the slug shown on the API page.
Error handling
Standard HTTP status codes + a stable `error.code` enum: `insufficient_funds`, `rate_limited`, `upstream_error`, `invalid_key`, `validation_failed`.
Usage metering
Every request returns an `x-apimarket-request-id` and `x-apimarket-cost-usd` header so you can reconcile usage client-side.
SSO / embed into your app
Hand your users off to OneAPIKey for sign-up + key provisioning in two clicks. See the SSO guide.

OpenAI-compatible endpoint

Base URL:https://oneapikey.app/v1

from openai import OpenAI
client = OpenAI(
    api_key=os.environ["ONEAPIKEY_API_KEY"],
    base_url="https://oneapikey.app/v1",
)
res = client.chat.completions.create(
    model="openai/gpt-5",
    messages=[{"role": "user", "content": "ship it"}],
)

Errors

All errors return JSON: { "error": { "code": "…", "message": "…" } }

Response headers

  • x-apimarket-request-id — trace a call in your dashboard.
  • x-apimarket-cost-usd — exact cost charged to your account.
  • x-apimarket-upstream-ms — upstream latency in milliseconds.

SSO (2-click connect)

  1. Register an SSO client in the admin console.
  2. Redirect your users to /sso/authorize?client_id=…&redirect_uri=…&state=…
  3. User approves, we issue a code. Exchange it server-side for an access token + provisioned key.