Connect your AI tools to OpenTokenMarket

OpenTokenMarket runs a shared LiteLLM gateway in front of every major model provider. When a sponsor funds your project you get one API key that works with every AI coding tool in this section — no provider accounts, no per-model keys, no billing setup.

What you get

• One OpenAI-compatible endpoint — https://v1.opentokenmarket.dev
• One stable key — your project access key. It never changes; sponsors just top up the budget behind it.
• Every model on the gateway — call them by the names listed in the model catalog.
• Usage and budget tracked for you — spend shows up on your project control panel, and the gateway stops requests when the funded budget runs out.

Where your key and URL come from

Both live on your project control panel (Dashboard → My Projects → your project):

• Project Access Key card — your key and the base URL, ready to copy.
• Code Snippet Generator — the same values wired into Node, Python, and cURL.

The card is empty until a sponsor funds the project. Until then you can still read these guides and prepare your config.

60-second quickstart

1. Open your project control panel and copy the Project Access Key and base URL.
2. Pick your tool below.
3. Paste the two values where that guide tells you.
4. Run a prompt. Watch the spend appear on your control panel.

Each guide is the same shape: what you get → prerequisites → configure → run → verify → troubleshooting. Pick yours from the list below.

Good to know

• Rate limits and budget are enforced at the gateway. A 429 or a budget has been exceeded error means the project's funded balance is used up — ask your sponsor to top up.
• The key is per-project. If you run several projects, each has its own key and budget.
• Keep the key server-side. Treat it like any secret; don't ship it in a browser bundle or commit it.
• Models can change. The model catalog is generated live from the gateway, so it always reflects what you can actually call today.