ShopOfToken is a storefront for large-model API access. Top up once, get a single API key, and call the models we have onboarded — billed by actual usage.
Gemini models are our current featured lineup; more models are being onboarded. The live list is shown in the console.
Google's Gemini models are fully onboarded, from flagship reasoning models to fast, cost-efficient ones for chat, multimodal and long-context work.
Additional leading models are being added. Your existing key and integration keep working as the catalog grows.
Every onboarded model shares the same base URL and authentication. Switching models is just changing the model name in your request.
Buying API access should feel like buying anything else online: simple and transparent.
Purchase, credit and key issuance happen online. You can make your first API call within minutes.
You are billed by actual usage per model. No separate account and payment setup for every provider.
Per-model prices are listed in the store and console, and every call is itemized in your usage log.
Create separate keys per project for clean usage tracking; revoke any key without touching the others.
Point your client at our base URL and authenticate with your key as a Bearer token. Minimal code changes.
Remaining credit and recent consumption are always visible in the console, so you never run dry mid-project.
From purchase to first successful call in minutes.
Choose a credit package in the store and check out. The credit lands in your account automatically.
Generate an API key in the console and store it somewhere safe — it is the credential for all your calls.
Send requests to https://api.chinatokenshop.com with Authorization: Bearer YOUR_API_KEY, using the official request format.
curl https://api.chinatokenshop.com/v1/chat/completions \
-H "Authorization: Bearer YOUR_API_KEY" \
-H "Content-Type: application/json" \
-d '{"model": "gemini", "messages": [{"role": "user", "content": "Hello"}]}'A storefront for large-model API access. We onboard leading models behind one gateway: one account, one API key, usage-based billing.
The Gemini family is our current lineup, with more models being onboarded. The console always shows the live, callable list.
Calls are charged against your credit by actual usage, at the per-model prices listed in the store and console.
Very little: set the base URL to our API endpoint and pass your key as a Bearer token. Request and response formats follow the official APIs.
Revoke it in the console immediately and issue a new one. Using separate keys per project keeps the blast radius small.