Gatelit Documentation

Gatelit is an AI gateway that sits between your applications and LLM providers. Any client — frontend, backend, or AI agent — makes authenticated LLM calls through a single endpoint. Provider routing, org-scoped keys, structured output, and request logging live in the gateway, not scattered across your codebase.

Quick Start

Make your first authenticated LLM request in minutes.

Authentication

Choose the right auth mode for your use case.

JavaScript SDK

Install and use the GatelitClient in any JS/TS project.

Vue SDK

Add reactive AI streaming to your Vue app with one composable.

How Gatelit works

Authenticate

Your backend issues a short-lived signed token (or use an OIDC token or a long-lived service key) that authorizes a specific model and user.

Send a request

Call POST /v1/chat/completions with the model in provider/model-name format. Gatelit routes to the right provider automatically.

Receive the response

Get back a standard OpenAI-compatible response — or stream it chunk by chunk. Token usage and cost are logged automatically.

Inspect in the dashboard

View request logs, manage provider keys, configure models, and edit saved prompts — all from the Gatelit dashboard.

Explore

Guides

Step-by-step walkthroughs for streaming, structured output, prompt management, and more.

Gateway API

Full reference for POST /v1/chat/completions and POST /v1/prompts/:id/run.

Dashboard

Manage keys, view logs, and control which models your org can use.

Bring Your Own Key

Use your own provider API keys scoped to your organization.