Getting started

Your first request in five minutes

Project Ares is one endpoint in front of Claude, GPT, and open models — metered, org-aware, and billed to one balance. If you've used any modern LLM API, you already know how this works.

1 · Create an account

Sign up at the API Platform Console with your email and verify it from your inbox. (When social sign-in options are enabled they appear right on the sign-in screen.) Every new account starts a free trial: trial credits valid for 30 days, no card required. When they run out or expire, pick a plan or buy credits to continue.

2 · Create an API key

In the console, open API keys → New key. Keys are shown once and stored hashed. A key is bound to the account that created it — your personal account, or an organization if you're working in one (that's whose credits it consumes).

3 · Call the API

The endpoint speaks the industry-standard Messages protocol, so the SDK you already use works by changing the base URL and key:

# curl curl https://api.projectares.ai/v1/messages \ -H "x-api-key: sk-ares-YOUR_KEY" \ -H "content-type: application/json" \ -d '{ "model": "claude-sonnet-5", "max_tokens": 1024, "messages": [{"role": "user", "content": "Hello, Project Ares"}] }'
# Python from anthropic import Anthropic client = Anthropic( base_url="https://api.projectares.ai", api_key="sk-ares-YOUR_KEY", ) msg = client.messages.create( model="claude-sonnet-5", # or "gpt-5", or an open model — same call max_tokens=1024, messages=[{"role": "user", "content": "Hello, Project Ares"}], )
Model IDs come from GET /catalog — the catalog is managed by Project Ares staff, so new models appear in your picker without code changes. The free ares-mock model streams a canned reply at $0 for pipeline testing.

4 · Watch it land

Every request is metered as it finishes: the Usage page shows cost per day, per model — and per member, if you're an organization admin. The Billing ledger shows the exact charge.