- One endpoint for multiple AI providers
- Unified request and response formats
- Transparent usage tracking and pricing
- Lightweight SDKs and modern developer experience
Why the Gateway exists
Working with multiple AI providers is powerful but messy - each has unique endpoints, model ids and types, and schemas.The Gateway removes that friction by providing a normalised API layer that automatically routes your request to the best available model or provider. You focus on building, not integration maintenance.
Core features
Unified Endpoints
Standardised chat, image, and moderation endpoints that work
consistently across all providers.
Smart Routing
Requests are dynamically routed to the optimal provider for speed, cost,
or reliability.
Built-in Usage Tracking
Every request is logged with latency, token counts, and cost metrics for
transparent reporting.
BYOK (Bring Your Own Key)
Optionally connect your own provider API keys to manage billing and rate
limits yourself whilst still benefiting from the Gateway’s unified
interface.
Getting started
- Sign up for an AI Stats account.
- Generate an API key in your dashboard.
- Make your first request to the
/v1/chat/completionsendpoint.
Quickstart
Jump into our quickstart to make your first successful API call in under a
minute.
Next steps
Authentication
Learn how to authenticate securely using your personal or team API key.
API Reference
Browse the complete API documentation with endpoints, parameters, and
response schemas.
Best Practices
Guidance on how to get the most out of the AI Stats Gateway.
Examples
Copy ready-to-run snippets for Next.js, Node.js streaming, Python, and
curl.
Need more detail? Continue with Integrating with the Gateway for an end-to-end walkthrough, then explore SDKs and Tools or Rate limits as you scale.