# Cursor in China — OpenAI-compatible, no VPN

> Markdown mirror of https://router.one/cursor-china for AI assistants and crawlers. Router One is an OpenAI-compatible LLM API gateway.
> Last updated: 2026-08-26

Router One connects Cursor to 40+ models from Mainland China without a VPN: override Cursor's OpenAI base URL with https://api.router.one/v1, use one Router One key for the chat models you pick, and top up with a card or Alipay — no foreign credit card required. Reaching the OpenAI API directly from the mainland is slow and unreliable and bills to a foreign card; the gateway removes both problems.

## Direct access, no VPN

Cursor talks to the OpenAI-compatible endpoint at https://api.router.one/v1. Review the published benchmark below before relying on any point-in-time network result.

## One key, 40+ models

Reach GPT, Claude, Gemini and more through a single OpenAI-compatible key. Pay-as-you-go pricing as low as 10% of official list prices on select models.

## Pay in RMB

Top up by card or Alipay through one hosted checkout, or with USDT/USDC on six chains. No foreign credit card required.

## Review the China latency benchmark

This setup guide does not make a separate latency promise. The published snapshot, last updated 2026-05-15, covers Beijing, Shanghai, and Shenzhen on China Telecom, China Unicom, and China Mobile in a rolling 7-day window. It is not an SLA, and individual networks may vary; review the full method and caveats before judging fit for your workload.

## 3-step setup

Override Cursor's OpenAI base URL in Settings, paste your Router One key, and pick a model. No VPN or proxy required. Cursor currently accepts a custom key only on its paid plans — Router One replaces your model bill, not your Cursor subscription. Setting up Cursor outside Mainland China? The global bring-your-own-key guide at /integrations/cursor walks through the same override.

### Create a Router One API key

Sign up at router.one and create an API key in the dashboard. Top up by card or Alipay on the hosted checkout, or with USDT/USDC — no foreign credit card required.

### Override the OpenAI base URL in Cursor

Open Cursor → Settings → Models → OpenAI API Key. Enable "Override OpenAI Base URL", set it to https://api.router.one/v1, and paste your Router One key.

### Verify and start coding

Click Verify, then use Cursor's chat with your chosen model — requests now route through Router One's China-reachable, OpenAI-compatible endpoint.

`Cursor settings`

```bash
# Cursor → Settings → Models → OpenAI API Key
Base URL: https://api.router.one/v1
API Key:  sk-your-router-one-key
```

- No low fixed cap for normal paid usage. Abuse prevention, per-account protection limits, and upstream constraints may still apply; if a request returns 429, check Dashboard -> Logs or contact support to raise limits.

## FAQ

### Do I need a VPN?

No proxy or VPN is required for the Router One endpoint. The benchmark below documents the dated network snapshot and its caveats.

### Which models can I use in Cursor?

Any chat-capable model Router One exposes — GPT, Claude, Gemini and more. Select a model in Cursor, or add the exact model ID under custom models if it is not listed. Your key covers the chat models you pick in the chat and Agent panel; Tab autocomplete, Auto, and Cursor's own Composer models stay on Cursor's built-in models.

### Why does Claude Opus 5 hit a context limit in Cursor?

Set Cursor's Context parameter for that model to 1M. Router One serves claude-opus-5 with a million-token window — the /models page shows the live figure, 1.05M today — but Cursor sends whatever Context value the model entry carries: open the model picker in the chat pane, edit the model's parameters, switch Context to 1M (older Cursor builds call this the MAX toggle), then start a new chat before you retry.

### How does billing work?

Use prepaid wallet balance or a live subscription plan. Top up by card or Alipay in one hosted checkout, or with USDT/USDC on six chains — usage outside plan coverage is metered per token.

### Is my code stored?

No. Router One proxies requests in real time; prompts and completions are not persisted. We log only metadata (model, tokens, latency, cost) for billing and observability.

## See also

- Manage Cursor alongside other coding tools: https://router.one/use-cases/ai-coding-tools
- Understand the OpenAI-compatible API: https://router.one/openai-compatible-api
- Fund the Cursor wallet with Alipay: https://router.one/alipay-llm-api
- Set up Codex CLI for terminal work: https://router.one/codex-china
- Bring your own key to Cursor anywhere: https://router.one/integrations/cursor
- Canonical page: https://router.one/cursor-china
- Models and per-model token rates: https://router.one/models (markdown: https://router.one/models.md)
- Pricing: https://router.one/pricing
- API docs (markdown): https://router.one/docs.md
- Company facts: https://router.one/facts/company.md