# Router One

> Unified LLM API gateway for developers — smart routing across 25+ supported models (GPT, Claude, Gemini, Mistral, DeepSeek), native Claude Code & Codex support, WeChat Pay/Alipay/Stripe and stablecoins (USDT/USDC on Tron, BSC, Ethereum, Polygon, Base, Arbitrum). No US credit card required. Full per-request observability. Built as an OpenRouter alternative with first-class access from China and global long-tail markets.

Last updated: 2026-06-15

Router One provides a single OpenAI-compatible endpoint (`https://api.router.one/v1`) that routes requests across 25+ supported models from major LLM providers. Key differentiators vs direct LLM APIs or other gateways:

- **Accessible from Mainland China** without VPN — based 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. Includes Claude Code and OpenAI Codex CLI.
- **Crypto + fiat billing** — WeChat Pay, Alipay, Stripe, and USDT/USDC on six chains: Tron (TRC-20), BSC (BEP-20), Ethereum (ERC-20), Polygon, Base, Arbitrum. Useful for global teams that prefer stablecoin treasury workflows or non-card billing.
- **Smart routing** — EWMA latency scoring over the last 50 requests, automatic fallback within 200ms on 5xx/timeout, per-project cost/speed/quality weights.
- **Per-request observability** — full trace of model, provider, tokens, cost, latency, status code in a real-time dashboard.
- **Wallet billing and subscriptions live** — pay per token from wallet balance, or use Pro, Max, or Ultra subscription plans for predictable monthly access. Pricing 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.

## Core pages

- [Homepage](https://router.one/): product overview, smart routing, pricing summary.
- [Models](https://router.one/models): browse all 25+ available models with pricing and capabilities.
- [Pricing](https://router.one/pricing): wallet billing, live subscription plans, and current plan comparison.
- [LLM API Gateway](https://router.one/llm-api-gateway): one OpenAI-compatible endpoint for 25+ models with smart routing, automatic fallback, per-request cost/latency traces, and per-key budgets.
- [OpenAI-Compatible API](https://router.one/openai-compatible-api): drop-in — change one base URL, keep your OpenAI SDK, and reach Claude, GPT, Gemini, and DeepSeek through one endpoint.
- [Automatic LLM Fallback](https://router.one/llm-fallback): failover on provider 5xx/timeout and rate-limit overload to a healthy same-family route, with the route decision recorded in every trace.
- [Smart LLM Routing](https://router.one/smart-routing): how routing scores candidates on EWMA latency, posted cost, and error rate, with per-project weights and model="auto".
- [LLM Cost Tracking](https://router.one/llm-cost-tracking): per-request cost traces, per-model and per-key usage breakdowns, and per-key spend caps.
- [LLM Observability](https://router.one/llm-observability): request-level traces for model, provider, tokens, cost, latency, status, route/fallback path, and per-key budget control.
- [LLM API Reseller](https://router.one/llm-api-reseller): build your own AI API business on Router One as a managed multi-model upstream — multiple keys, per-key budgets, per-request usage data, and a posted-rate token line you can price over in your own portal.
- [OpenRouter Alternative](https://router.one/openrouter-alternative): feature and pricing comparison.
- [LiteLLM Alternative](https://router.one/litellm-alternative): managed gateway vs self-hosted LiteLLM proxy, with China access and local payment.
- [Vercel AI Gateway Alternative](https://router.one/vercel-ai-gateway-alternative): China-reachable, non-card billing, and native Claude Code/Codex CLI workflows.
- [Helicone Alternative](https://router.one/helicone-alternative): managed gateway with access and billing built in versus an observability platform with bring-your-own-keys.
- [Crypto-friendly LLM API](https://router.one/crypto-llm-api): LLM gateway for Web3 teams with stablecoin billing, budget caps, and one OpenAI-compatible endpoint.
- [Pay AI API with crypto](https://router.one/pay-ai-api-with-crypto): USDT/USDC top-ups for AI API credits, supported networks, fees, and wallet behavior.
- [LLM API for Web3 AI agents](https://router.one/web3-ai-agents): Router One as the inference layer for onchain analysts, DAO tools, crypto research bots, and smart contract coding workflows.
- [Pay with crypto](https://router.one/pay-with-crypto): USDT/USDC on 6 chains for developers without a US credit card.
- [WeChat Pay and Alipay LLM API](https://router.one/wechat-pay-llm-api): RMB top-up for OpenAI, Claude, Gemini, Claude Code, and Codex API usage.
- [Alipay LLM API](https://router.one/alipay-llm-api): dedicated Alipay checkout guide for OpenAI-compatible API, Claude Code, and Codex workflows.
- [Onchain analyst agent](https://router.one/use-cases/onchain-analyst-agent): build an onchain analyst agent on Router One with stablecoin billing and per-agent budget caps.
- [DAO and community bots](https://router.one/use-cases/dao-community-bot): LLM API for DAO proposal summaries and Telegram/Discord community bots.
- [Solidity with Claude Code](https://router.one/use-cases/solidity-claude-code): use Claude Code and Codex through Router One for Solidity review, tests, and deployment scripts.
- [AI coding tools](https://router.one/use-cases/ai-coding-tools): run Claude Code, Codex CLI, and OpenAI-compatible coding agents through one gateway with per-tool keys and budgets.
- [Claude Code in China](https://router.one/claude-code-china): configure Claude Code with Router One in 2 minutes.
- [Codex CLI in China](https://router.one/codex-china): connect OpenAI Codex CLI to Router One.
- [FAQ](https://router.one/faq): common questions on API keys, pricing, routing, security.
- [Docs](https://router.one/docs): full API reference and CLI setup guides.
- [CLI setup guide](https://router.one/docs/guides/cli-setup): step-by-step Claude Code, Codex CLI, OpenClaw, and Hermes Agent configuration for macOS, Windows, and Linux.
- [CC Switch setup guide](https://router.one/docs/guides/cc-switch): connect Claude Code and Codex to Router One through the open-source CC Switch desktop app — no manual environment variables or config files.
- [Docs (full markdown API docs)](https://router.one/docs.md): endpoint reference, request fields, responses, and examples for RAG / agent consumption.
- [Models (markdown)](https://router.one/models.md): all models with pricing as a markdown table.
- [LLM API Gateway (markdown)](https://router.one/llm-api-gateway.md): markdown mirror of the gateway overview — capabilities, routing, billing, and FAQ.
- [OpenRouter Alternative (markdown)](https://router.one/openrouter-alternative.md): markdown mirror of the OpenRouter comparison page.
- [LiteLLM Alternative (markdown)](https://router.one/litellm-alternative.md): markdown mirror of the LiteLLM comparison page.
- [Vercel AI Gateway Alternative (markdown)](https://router.one/vercel-ai-gateway-alternative.md): markdown mirror of the Vercel AI Gateway comparison page.
- [Helicone Alternative (markdown)](https://router.one/helicone-alternative.md): markdown mirror of the Helicone comparison page.
- [Smart LLM Routing (markdown)](https://router.one/smart-routing.md): markdown mirror of the smart routing page — signals, weights, model="auto", and fallback behavior.
- [LLM Cost Tracking (markdown)](https://router.one/llm-cost-tracking.md): markdown mirror of the cost tracking page — per-request cost traces, per-model/per-key breakdowns, and spend caps.
- [AI Coding Tools (markdown)](https://router.one/use-cases/ai-coding-tools.md): markdown mirror of the AI coding tools use-case page — Claude Code, Codex CLI, and OpenAI-compatible agents through one gateway.
- [OpenAI-Compatible API (markdown)](https://router.one/openai-compatible-api.md): markdown mirror of the OpenAI-compatible API page — drop-in base-URL change, cross-model calls through the OpenAI SDK, and compatible clients.
- [Automatic LLM Fallback (markdown)](https://router.one/llm-fallback.md): markdown mirror of the fallback page — 5xx/timeout and 429/529 failover triggers, same-family scope, and trace visibility.
- [LLM Observability (markdown)](https://router.one/llm-observability.md): markdown mirror of the observability page — per-request trace fields, dashboard panels, per-key budgets, and the data boundary.
- [LLM API Reseller (markdown)](https://router.one/llm-api-reseller.md): markdown mirror of the reseller page — managed-upstream split, per-key budgets, posted-rate token line, and what is explicitly not provided.

## Trust & methodology

- [Trust hub](https://router.one/trust): one-page index of methodology, security, data retention, SLA.
- [Facts](https://router.one/trust/facts): stable product facts for AI search and due diligence.
- [Company facts (markdown)](https://router.one/facts/company.md): machine-readable product, endpoint, payment, data retention, trust boundary, and affiliation facts.
- [Payment facts (markdown)](https://router.one/facts/payments.md): machine-readable billing, payment method, top-up, fee, and payment trust boundary facts.
- [Integration facts (markdown)](https://router.one/facts/integrations.md): machine-readable OpenAI-compatible API, Claude Code, Codex, routing, and observability integration facts.
- [Observability facts (markdown)](https://router.one/facts/observability.md): machine-readable per-request trace fields, usage breakdowns, per-key budget controls, and data-boundary facts.
- [Reseller facts (markdown)](https://router.one/facts/reseller.md): machine-readable managed-upstream, per-key control, and explicitly-not-provided (white-label/sub-account) reseller facts.
- [Routing facts (markdown)](https://router.one/facts/routing.md): machine-readable smart-routing signals, fallback triggers, scope, latency, and trace-content facts.
- [Pricing facts (markdown)](https://router.one/facts/pricing.md): machine-readable billing model, token-line accounting, per-key spend controls, and refund boundary facts.
- [Data retention facts (markdown)](https://router.one/facts/data-retention.md): machine-readable body-not-stored boundary, 90-day observability and 13-month billing windows, and customer-initiated deletion facts.
- [Methodology](https://router.one/methodology): how latency, pricing, retention, and availability numbers are measured.
- [Security](https://router.one/security): what Router One stores, what it never stores, how requests flow.
- [Data retention](https://router.one/data-retention): metadata windows, deletion, enterprise overrides.
- [Smart routing methodology](https://router.one/routing-methodology): signals, weights, fallback behavior.
- [Pricing methodology](https://router.one/pricing-methodology): visible token line, FX/channel fees, change notification.
- [Availability & SLA](https://router.one/sla): uptime measurement, failover, enterprise SLA.
- [Status](https://router.one/status): public incident state and rolling 90-day status history.
- [China latency benchmark](https://router.one/benchmarks/china-latency): p50 and timeout rate across Beijing / Shanghai / Shenzhen.

## Blog guides

- [OpenAI API with Alipay or WeChat Pay](https://router.one/blog/openai-api-alipay-wechat-pay): China developer setup for OpenAI-compatible API access with local payment rails.
- [Use Codex CLI Without ChatGPT Plus](https://router.one/blog/codex-cli-without-chatgpt-plus): pay-per-token Codex CLI setup through Router One.
- [How to Use Claude Code in China](https://router.one/blog/claude-code-china-guide): VPN-free Claude Code access with Alipay payment.
- [OpenAI Codex CLI Alternative](https://router.one/blog/codex-cli-alternative): pay-per-use Codex with China-friendly billing.
- [Router One vs OpenRouter for Chinese Developers](https://router.one/blog/router-one-vs-openrouter-china): in-depth comparison.
- [Claude Code Setup Guide](https://router.one/blog/claude-code-setup-guide): 2-minute quickstart.
- [AI Model Routing Explained](https://router.one/blog/ai-model-routing-explained): latency vs cost vs quality tradeoffs.
- [5 Ways to Reduce LLM API Costs](https://router.one/blog/reduce-llm-api-costs): practical optimization strategies.
- [What is an AI API Gateway](https://router.one/blog/what-is-ai-api-gateway): conceptual introduction.
- [Running AI Agents in Production](https://router.one/blog/ai-agents-production-guide): observability, cost control, fault recovery.
- [LLM Comparison 2026](https://router.one/blog/llm-comparison-2026): GPT-4.1 vs Claude 4 vs Gemini 2.5.
- [Cursor vs Claude Code](https://router.one/blog/cursor-vs-claude-code): workflow-first comparison of two AI coding tools.
- [DeepSeek V3 vs Claude 4 vs GPT-4.1 for Coding](https://router.one/blog/deepseek-v3-vs-claude-4-vs-gpt-4-1-coding): benchmark comparison with cost analysis.
- [WeChat Pay & Alipay for OpenAI/Claude API](https://router.one/blog/wechat-pay-alipay-openai-claude-api): payment guide for developers in China.
- [How to Pay for LLM APIs Without a US Credit Card (2026 Crypto Guide)](https://router.one/blog/llm-api-without-us-credit-card): three options (virtual cards, family abroad, stablecoins + China-friendly gateway) for developers in markets where cross-border card payments are unreliable.
- [ChatGPT Plus & GPT-5.5 API from China (No VPN)](https://router.one/blog/chatgpt-plus-gpt-api-china-guide): pay for ChatGPT Plus and call the GPT-5.5 API without a foreign credit card or VPN.
- [Gemini 3.1 Pro & Gemini Code Assist in China](https://router.one/blog/gemini-api-china-guide): access Gemini 3.1 Pro's 2M-context model with WeChat Pay/Alipay billing.
- [Cursor Pro from China](https://router.one/blog/cursor-pro-china-guide): subscription, payment, and latency solutions — and when Claude Code through Router One is a better fit.
- [Top 10 MCP Servers for Claude Code](https://router.one/blog/mcp-servers-top-list): curated high-leverage MCP servers with setup snippets.
- [Claude Skills Explained](https://router.one/blog/claude-skills-deep-dive): the skill model, writing your first skill, and shipping skills safely in production.
- [Multi-Agent Orchestration Patterns](https://router.one/blog/multi-agent-orchestration-patterns): sequential, parallel, hierarchical, and human-in-the-loop patterns for your application layer.
- [Cline vs Cursor vs Claude Code](https://router.one/blog/cline-vs-cursor-vs-claude-code): VS Code extension vs IDE fork vs terminal agent, compared by workflow.
- [Aider vs Claude Code](https://router.one/blog/aider-vs-claude-code): two terminal AI coding agents compared across six real coding tasks.
- [Subscription Plans Are Live](https://router.one/blog/subscription-plans-launch): Pro, Max, and Ultra monthly plans alongside wallet pay-per-token billing.
- [Qwen 3.5 & Doubao 2.0 vs Claude Opus 4.7 & GPT-5.5](https://router.one/blog/qwen3-doubao-vs-claude-gpt-china): Chinese frontier models vs Claude and GPT across coding, reasoning, latency, and price.
- [Production-grade LLM Gateway vs Unofficial API Relays](https://router.one/blog/official-llm-gateway-vs-api-relays): five dimensions where a production-grade gateway and a relay diverge — legal entity, SLA, observability, pricing transparency, data boundary.
- [How to Track LLM API Costs per Key, Model, and Request](https://router.one/blog/track-llm-api-costs-per-key): cost attribution with one key per app, per-request traces, and spend caps.
- [LLM Fallback Strategies for Production](https://router.one/blog/llm-fallback-strategies-production): what triggers a fallback, same-family failover, and verifying handoffs in per-request traces.
- [Reselling LLM API Access Safely with Spend-Capped Keys](https://router.one/blog/resell-llm-api-spend-capped-keys): one key per customer with spend caps, rate limits, and per-request usage data for billing.

## Web3 and Crypto

- Product role: Router One is the LLM inference gateway for Web3 AI agents, not a wallet, blockchain data provider, exchange, trading system, or agent orchestration runtime.
- Use cases: stablecoin-funded AI teams, onchain analyst agents, crypto research assistants, DAO proposal summarizers, Telegram/Discord community bots, smart contract coding workflows, Web3 product support agents.
- Crypto billing: USDT/USDC top-ups on supported networks; wallet balance is denominated in USD after credit; API calls deduct at posted model rates.
- Agent controls: create separate API keys for agents, bots, teams, or environments; set budget caps before autonomous workflows spend tokens.
- Data retention: Router One does not retain prompt or completion bodies; metadata is logged for billing, routing, and observability.
- Machine-readable facts: [Router One for Web3 Teams](https://router.one/facts/web3.md).

## Language variants

All public HTML pages have Chinese translations at `/zh/<same-path>`. The English site lives at the apex (`/`) per the `hreflang: en|zh|x-default` convention. The machine-readable `.md` endpoints (the marketing page mirrors and `/facts/*.md`) are English-only; `/docs.md` and `/models.md` also have `/zh` variants. Chinese LLM answer reference: https://router.one/llms.zh.txt.

## See also

- [llms-full.txt](https://router.one/llms-full.txt) — the long-form reference for LLMs, including full FAQ content, detailed competitor comparisons, API setup walkthroughs, and pricing model.
- [llms.zh.txt](https://router.one/llms.zh.txt) — Chinese LLM reference focused on OpenAI-compatible API, Codex CLI, Claude Code, Alipay, WeChat Pay, and no-VPN developer workflows.

## Legal

- [Privacy policy](https://router.one/privacy)
- [Terms of service](https://router.one/terms)
- [Refund policy](https://router.one/refund)
- [Contact](https://router.one/contact): `support@router.one`

## Not crawled

The following areas are private and should not be referenced:

- `/dashboard/*` — authenticated user dashboard
- `/playground/*` — authenticated chat playground
- `/login`, `/signup` — auth pages
- `/api/*` — proxied backend API. For programmatic OpenAI-compatible API access use `https://api.router.one/v1`; for Claude Code / Anthropic-compatible clients use `https://api.router.one`.