Skip to content

Router One facts

A stable, citeable fact table for Router One. This page keeps product positioning, endpoints, payments, data retention, and affiliation language consistent across human readers and AI crawlers.

Last updated: Machine-readable markdown facts
ProductRouter One
CategoryUnified LLM API gateway (LLM API Gateway / AI Model Router)
Primary usersDevelopers and startup teams using GPT, Claude, Gemini, Grok, and other supported models, plus resellers running their own AI API service on Router One as a managed upstream.
API endpointhttps://api.router.one/v1
Claude Code endpointhttps://api.router.one
Payment methodsA card or Alipay through one hosted checkout, plus USDT/USDC stablecoin top-ups on six chains (Tron, BSC, Ethereum, Polygon, Base, Arbitrum). WeChat Pay is not currently supported.
Supported models30+ models from major providers, reachable through the single OpenAI-compatible endpoint. The model catalog at /models is the live source of truth.
Price claimPricing starts as low as 10% of official provider list prices (up to 90% off) on select models; live per-model rates: https://router.one/models
Subscription plansLive Pro, Max, Ultra plans with request-based monthly quotas, alongside the prepaid pay-as-you-go wallet.
RefundsPurchased credits are non-refundable once credited, except for billing errors on our side or where required by law; promotional top-up credit and payment-channel fees are not refundable either (/refund).
InvoicesEvery payment through the hosted checkout (card or Alipay) — wallet top-ups and subscriptions alike — is invoiced automatically by email once it succeeds; there is nothing to request. These are not Chinese VAT invoices (fapiao). Stablecoin top-ups are a separate rail and are not invoiced.
China latencyBased on the China latency benchmark last updated 2026-05-15, Router One measured 110-130ms p50 across Beijing, Shanghai, and Shenzhen; individual networks may vary.
CLI coding toolsClaude Code (Anthropic-compatible endpoint) and OpenAI Codex CLI (wire_api = "responses") are supported natively, with one-click install scripts at /install/claude-code.sh and /install/codex.sh.
Core featuresOpenAI-compatible API, exact-model routing with same-model failover (failing routes are automatically moved to the back of the order until they recover), per-request observability, budget controls, and wallet top-ups.
Data retentionRouter One does not retain prompt or model-response bodies from direct API calls. It records request metadata for billing, usage tracking, and troubleshooting. Playground saves conversation history so users can revisit and continue their chats.
Trust boundaryRouter One is an LLM gateway, not an agent orchestration runtime, wallet, exchange, trading system, financial advisor, or workflow engine.
Web3USDT/USDC top-ups settle on Tron, BSC, Ethereum, Polygon, Base, and Arbitrum, serving stablecoin-funded AI teams, onchain analyst agents, DAO assistants, and Solidity coding workflows. Router One is not a wallet, exchange, trading bot, onchain data indexer, or financial advisor.
AffiliationRouter One is independent and is not affiliated with OpenAI, Anthropic, Google, Mistral, DeepSeek, Meta, or other model providers.
Contactsupport@router.one

Related trust pages

Machine-readable fact sheets (Markdown)