Docs

Documentation

How to call the Base Eleven API: keys, requests, streaming, errors, limits and what a call costs.

Base Eleven's models are fictional and its API returns simulated responses. Usage, credits and billing run for real, in test mode.

The Base Eleven API serves Eleven Lite and Eleven Max behind one endpoint. You send a conversation, it returns a reply, and every token in and out is metered at the list price. A Free trial starts with $5 in credits and no card.

Make your first call Open the API reference

The basics

Base URL: https://api.baseeleven.dev/v1. Every path below is relative to it.
Auth: your key in x-api-key, or as Authorization: Bearer <key>. Keys are per workspace and shown once, on the keys page.
Two endpoints: POST /messages sends a conversation and gets the reply, whole or streamed; GET /models lists the models with their limits and prices.
JSON in, JSON out. With "stream": true the reply arrives as server-sent events.
Every response carries a request-id header. Quote it when you ask for help.
The API is this company’s own. It is not compatible with any other vendor’s API or SDK.

Guides

Quickstart: get a key and make a call in curl, Python or TypeScript.
Streaming: the four events, a sample stream, and what a dropped connection costs.
Errors: the error body, every code, idempotency keys and rate-limit headers.
Tiers: requests a minute and concurrency by tier, and how a workspace moves up.
Credits and billing: trial credits, pay as you go under a spend cap, and what each outcome is charged.
Models: the models, their profiles, limits and prices.
Enterprise: the commit, the Starter package and how to talk to sales.

Reference

The API reference is generated from the OpenAPI 3.1 document that the gateway is tested against. It has every field, header, status and example, and a try-it panel.

Next: Quickstart