Grok API in China — no VPN, no foreign card

Router One makes Grok chat models callable from Mainland China without a VPN or foreign card: the same OpenAI-compatible endpoint https://api.router.one/v1 and key that reach GPT, Claude and Gemini. Each request the gateway sends to a model is logged with the wallet charge or plan quota it used (a streamed request that failed before any output may have no record). xAI's own endpoints are blocked for mainland networks and bill to a foreign card; the gateway removes both blockers.

Get your API key

Direct access, no VPN

Grok models answer through the OpenAI-compatible endpoint at https://api.router.one/v1. Review the published benchmark below before relying on any point-in-time network result.

Token pricing with a ledger

Grok chat models are metered per token at Router One's posted rates. As of the 2026-11-05 plan response, grok-4.7, grok-4.6 and grok-4.5 are also in the Mid-tier models tier of the Pro, Max and Ultra plans, so on a plan their requests draw on that allowance first; grok-4.7-build-fast is in no plan tier and bills to the wallet. The site-wide up to 90% off official prices figure applies to select models across the catalog; discount levels vary by model, and the /models catalog shows each Grok model's current rate against its base rate (some Grok models carry no discount). Every request the gateway sends to a model logs its model, tokens, latency, and the wallet charge or plan quota it used in Dashboard → Logs (a streamed request that failed before any output may have no record).

Pay in RMB

Top up with a card or Alipay through one hosted checkout, or with USDT/USDC on six chains. Metered per token, pay-as-you-go — no xAI account, no foreign card.

Review the China latency benchmark

This page 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.

Read the published benchmark

Grok chat models in the catalog

The lineup follows the live catalog — current per-token rates are on the models page.

Model ID	What it is
grok-4.7	Grok 4.7 chat model (listed 2026-09-22) — 500k context, text and image input, vision and tool calling; Chat Completions and native Responses
grok-4.7-build-fast	Grok 4.7 Build Fast (listed 2026-09-23) — 500k context, text and image input, vision and tool calling; a separate id with its own rate line on the model page
grok-4.6	Previous Grok chat model — 500k context, vision and tool calling
grok-4.5	Earlier flagship — 500k context, vision and tool calling

See live Grok model pricing Gemini API in China on the same key

Grok Imagine on the same key

Grok Imagine is in the catalog as two entries — grok-imagine-image and grok-imagine-image-quality — both at a flat per-image unit price, served at the standard images endpoint with the same key and wallet.

Image generation API guide

Grok Video is not currently in the catalog, and no other video model is listed either (none since 2026-09-05). The video page covers video generation availability from China — check /models before planning a clip.

Video generation availability in China

3-step setup

Any OpenAI-compatible client works — set the base URL, pick a Grok model ID from the catalog, and call.

1Create a Router One API keySign up at router.one and create an API key in the dashboard. Top up with a card or Alipay on the hosted checkout, or with USDT/USDC — no foreign credit card required.
2Point your client at the gatewaySet the base URL to https://api.router.one/v1 in the OpenAI SDK or any compatible tool, and use a Grok model ID copied from the /models page.
3Send a request and verifyMake a chat completion call, then open Dashboard → Logs to see the trace: model, tokens, latency, and the wallet charge or plan quota the request used.

grok-via-router-one.sh

# Grok via the OpenAI-compatible endpoint
curl -X POST https://api.router.one/v1/chat/completions \
  -H "Authorization: Bearer sk-your-router-one-key" \
  -H "Content-Type: application/json" \
  -d '{"model": "<grok-model-id>",
       "messages": [{"role": "user", "content": "你好"}]}'

No low fixed cap for normal paid usage. Each key and each account runs under default request-per-minute and token-per-minute limits, and upstream constraints may still apply; if a request returns 429, back off and retry (honor Retry-After when it is sent), and if your traffic needs higher limits, email support@router.one to have the key or account limits raised.

FAQ

Why can't I call the Grok API directly from China?

xAI's API endpoints are blocked for Mainland China network space, and billing requires a foreign card. A VPN adds latency and instability; a China-reachable gateway avoids both.

Which Grok models are available?

The chat lineup currently spans grok-4.7 (listed 2026-09-22), grok-4.7-build-fast (listed 2026-09-23), grok-4.6 and grok-4.5. As of the 2026-10-19 catalog, Router One no longer lists Grok 4.3 (grok-4.3) or Grok Build 0.1 (grok-build-0.1). Grok 4.7 and Grok 4.7 Build Fast are covered in the September 2026 new-model guide. The catalog updates as models ship or retire, so treat /models as the source of truth for the live lineup, per-token rates, context windows, and capability tags.

Do I use xAI's SDK?

No — you call Grok models through the OpenAI-compatible interface: the official OpenAI SDK, curl, or any compatible client. Vision input on grok-4.7, grok-4.7-build-fast, grok-4.6 and grok-4.5 works via standard image_url content parts; grok-4.7 and grok-4.7-build-fast are the Grok chat ids whose catalog entries list image among their input modalities. Every Grok chat id is also served natively on /v1/responses (since 2026-09-22), so Codex CLI can point at any of them.

How does pricing compare to official rates?

Grok chat models are billed per token at Router One's posted rates. The site-wide "up to 90% off official prices" figure is a ceiling reached on select models across the catalog, not a Grok-specific rate — open the Grok rows on /models to see each model's current discounted rate against its base rate (some Grok models carry no discount).

Do Router One plans cover Grok?

As of the 2026-11-05 plan response, Router One's Pro, Max and Ultra plans cover three of the four Grok chat ids: grok-4.7, grok-4.6 and grok-4.5 are in the Mid-tier models tier of all three, so their requests use that tier's allowance first and bill the wallet once it runs out, while grok-4.7-build-fast, a separate id, is in no plan tier and bills the wallet per token. Dashboard → Logs shows, for each request, the wallet charge or the plan quota it used.

Can the same key generate Grok images?

Yes — Grok Imagine is listed as grok-imagine-image and grok-imagine-image-quality, both unit-priced per image on the same key and wallet. Grok Video is not currently in the catalog, and no video model has been listed since 2026-09-05; the image generation API page covers images, and the video generation page covers video availability.

Gemini API in China Video generation API in China Image generation API Pricing methodology Top up with Alipay Grok 4.7 vs Grok 4.6 Grok 4.6 vs Grok 4.5 Grok 4.7 and Grok 4.7 Build Fast in the September 2026 model guide

Start calling Grok from China

Get your API key