# Tokios

## Docs

- [Use Claude Code with Your Own Local Models via Tokios](https://docs.tokios.com/guides/claude-code.md): Point Claude Code at your private local model by setting ANTHROPIC_BASE_URL and ANTHROPIC_API_KEY to route every request through the Tokios gateway.
- [Connect Any OpenAI or Anthropic-Compatible Client to Tokios](https://docs.tokios.com/guides/compatible-clients.md): Any client that supports a custom base URL for OpenAI or Anthropic APIs can route to your local model through Tokios with two config changes.
- [Route Requests to Multiple Local Models with Tokios](https://docs.tokios.com/guides/model-routing.md): Register multiple local model deployments under different names and switch between them by changing the model field in your API request.
- [Use OpenAI Codex with Your Own Local Models via Tokios](https://docs.tokios.com/guides/openai-codex.md): Route OpenAI Codex CLI requests to your private local model by setting OPENAI_BASE_URL and OPENAI_API_KEY to point at the Tokios gateway.
- [How Tokios Routes Requests to Your Local Models](https://docs.tokios.com/how-it-works.md): Learn how Tokios uses an outbound-only WebSocket tunnel to route API requests to your local LLMs without exposing inbound ports or a public IP address.
- [Tokios: Private Local Models Behind One API Endpoint](https://docs.tokios.com/index.md): Tokios tunnels requests to your local LLMs through an outbound-only WebSocket, exposing them via a single OpenAI- and Anthropic-compatible API endpoint.
- [Get Started with Tokios in Under 5 Minutes](https://docs.tokios.com/quickstart.md): Install the Tokios Connector, pair your local model, and make your first OpenAI-compatible API call in under five minutes — no port forwarding needed.
- [Tokios API Endpoints: OpenAI & Anthropic Reference](https://docs.tokios.com/reference/api-endpoints.md): Complete reference for Tokios API endpoints including POST /v1/chat/completions, /v1/messages, and /v1/responses with request and response schemas.
- [Tokios Connector Configuration: tokios.json Reference](https://docs.tokios.com/reference/connector-config.md): Full reference for the tokios.json configuration file and connector CLI flags, including tunnel_token, upstream, and log-level options.
- [Tokios FAQ: Setup, Connectivity, and Troubleshooting](https://docs.tokios.com/reference/faq.md): Answers to common questions about Tokios setup, firewall requirements, API compatibility, streaming, data privacy, supported backends, and billing.
- [Tokios API Keys: Create, Use, and Revoke Bearer Tokens](https://docs.tokios.com/setup/api-keys.md): Generate tenant-scoped API keys for authenticating requests to api.tokios.com. Keys start with sk-tokios- and are passed as Bearer tokens.
- [Download, Install, and Configure the Tokios Connector](https://docs.tokios.com/setup/connector-install.md): Download the tokios-connector binary for Windows, macOS, or Linux, configure tokios.json, and establish your local tunnel in minutes.
- [Register a Model Deployment in the Tokios Dashboard](https://docs.tokios.com/setup/register-model.md): Pair the Tokios Connector with your dashboard account to register your local model as a named deployment reachable at api.tokios.com.
- [Supported Local Model Backends — Tokios Connector](https://docs.tokios.com/setup/supported-backends.md): Configure the Tokios Connector to forward traffic to Ollama, llama.cpp, vLLM, or LM Studio by setting the correct upstream URL in tokios.json.