> ## Documentation Index
> Fetch the complete documentation index at: https://docs.muxx.dev/llms.txt
> Use this file to discover all available pages before exploring further.

# Gateway Overview

> Route your LLM requests through Muxx Gateway for logging, caching, and rate limiting.

The Muxx Gateway is a proxy that sits between your application and LLM providers. It intercepts requests, logs them, and can apply caching and rate limiting—all with minimal code changes.

## How It Works

```
Your App → Muxx Gateway → LLM Provider (OpenAI, Anthropic, etc.)
              ↓
         Logging & Metrics
              ↓
         Muxx Dashboard
```

1. Your app sends requests to the Muxx Gateway instead of directly to the provider
2. The gateway forwards requests to the appropriate provider
3. Responses are logged and returned to your app
4. View everything in the Muxx dashboard

## Key Features

<CardGroup cols={2}>
  <Card title="Request Logging" icon="list">
    Every request and response is logged with full payloads, token counts, and timing.
  </Card>

  <Card title="Response Caching" icon="database">
    Cache identical requests to reduce latency and costs.
  </Card>

  <Card title="Rate Limiting" icon="gauge">
    Protect your budget with requests-per-minute limits.
  </Card>

  <Card title="Cost Tracking" icon="dollar-sign">
    Real-time cost calculation for every request.
  </Card>
</CardGroup>

## Supported Endpoints

The gateway supports OpenAI-compatible and Anthropic-compatible endpoints:

| Provider  | Endpoint               | Gateway URL                                    |
| --------- | ---------------------- | ---------------------------------------------- |
| OpenAI    | `/v1/chat/completions` | `https://gateway.muxx.dev/v1/chat/completions` |
| Anthropic | `/v1/messages`         | `https://gateway.muxx.dev/v1/messages`         |

## When to Use Gateway

The gateway is ideal when you want:

* **Zero code changes** - Just update the base URL
* **Caching** - Reduce costs and latency for repeated queries
* **Rate limiting** - Protect against runaway costs
* **Quick integration** - Get logging in minutes

For deeper tracing with spans and custom instrumentation, combine the gateway with the [SDK](/sdk/python/quickstart).

## Next Steps

<Card title="Gateway Setup" icon="rocket" href="/gateway/setup">
  Configure the gateway for your application
</Card>