Developer documentation

The Futureway AI API

An OpenAI-compatible inference API on EU-hosted models and an App API for the platform itself — one key, one base URL, the official SDKs.

Quickstart

Create a key, set the base URL, make your first call — in five minutes.

Authentication

Personal and service keys, scopes, expiry, security rules.

API reference

Every endpoint with parameters, examples and error codes.

SDKs & tools

OpenAI SDKs, OpenCode, Continue, LangChain, Vercel AI SDK.

Two surfaces, one key

The inference API speaks the OpenAI format — every OpenAI SDK works with a changed base URL. The App API is our own surface for what the platform can do: key introspection today, company knowledge and agents next.

Inference API: https://api.futureway.ai/v1
App API: https://api.futureway.ai/v1
Authentication: Authorization: Bearer $FUTUREWAY_API_KEY
OpenAPI 3.1: /developers/openapi.json

Both surfaces share the base URL and accept the same API key. Browser requests are rejected — the API is built for server-to-server calls.

Your first call

Usage is billed from your organisation's prepaid balance. Create a key under Settings → API & CLI, export it as an environment variable and call futureway-smart:

curl https://api.futureway.ai/v1/chat/completions \
  -H "Authorization: Bearer $FUTUREWAY_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
  "model": "futureway-smart",
  "messages": [
    {
      "role": "user",
      "content": "Was ist die Hauptstadt von Frankreich?"
    }
  ]
}'

Endpoints

Data processing

Details on EU sovereignty.