Frequently Asked Questions
Everything you need to know about using Router One. Can't find what you're looking for? Feel free to contact us.
What is Router One?
Router One is an LLM API gateway. It provides a unified OpenAI-compatible API endpoint for 40+ supported models across the major model families (GPT, Claude, Gemini, and more), with intelligent routing, automatic fallback, cost optimization, and per-request observability — reachable worldwide, including mainland China, with payment by card, Alipay, or stablecoins.
How do I get an API Key?
Sign up for a Router One account, then go to Dashboard → API Keys to create a new key. Your key will start with "sk-". Use it in the Authorization header as "Bearer sk-xxx".
Which models are supported?
The line-up turns over whenever new model versions ship, so this answer does not pin a fixed list. The model catalog at /models is the live source of truth: it lists every model ID you can call right now — the GPT, Claude and Gemini families among others — with per-token rates, context windows, and capabilities. If a model is not in the catalog, it is not callable through the gateway.
How does smart routing work?
Your request names a model and the gateway selects an eligible provider route. Server-side adaptive paths can use recent latency, posted cost, and reliability signals; retryable upstream failures can move to another healthy candidate, with the final route recorded in the request trace. The public dashboard does not expose project-level routing weights.
How is billing handled?
Router One supports pay-per-token wallet billing and live subscription plans. Top up your balance with a card or Alipay through one hosted checkout, or with USDT/USDC stablecoins on six chains, or choose a Pro, Max, or Ultra plan for predictable monthly access. If your usable credit reaches zero, API calls outside plan coverage will return a 402 error — top up to resume.
Is my data safe? Do you store request content?
We process requests in real time and do not retain your prompts or completions. Each request is proxied directly to the upstream provider. We only log metadata (model, tokens, cost, latency) for billing and observability — never the actual content.
How do I configure Claude Code or Codex to use Router One?
For OpenAI-compatible SDKs and HTTP clients, set the base URL (or OPENAI_BASE_URL) to https://api.router.one/v1. For Claude Code, set ANTHROPIC_BASE_URL to https://api.router.one and ANTHROPIC_AUTH_TOKEN to your Router One key — ANTHROPIC_API_KEY is not required, and setting it triggers an extra approval prompt. For Codex CLI, add a provider in ~/.codex/config.toml with base_url = "https://api.router.one/v1", wire_api = "responses", and env_key = "ROUTER_ONE_API_KEY" — Codex does not read OPENAI_BASE_URL or OPENAI_API_KEY for a custom provider. One-click install scripts and detailed setup guides are in our documentation.
What should I do if I encounter an issue?
Check the Dashboard → Logs page to inspect recent requests and error codes. Common issues: 401 (invalid API key), 402 (insufficient balance), 429 (rate limit exceeded). For further help, email support@router.one or open an issue on our GitHub.
Can I get an invoice?
Router One issues invoices for any top-up completed through the hosted checkout — card or Alipay — including plain pay-as-you-go wallet top-ups, not only enterprise or subscription accounts. Email support@router.one with your account ID to request one. Stablecoin top-ups are a separate rail and are not invoiceable.