Chuizi.AI
Chuizi.AI

Overview

Chuizi.AI is a unified AI API gateway that lets you access 200+ models from 17 providers through a single API key and baseURL. Just swap your baseURL and API key — your existing code works instantly, zero changes needed.

Why Chuizi.AI

Managing multiple AI provider accounts, API keys, and SDKs is painful. Each provider has its own authentication, rate limits, billing, and failure modes. Chuizi.AI consolidates all of this behind a single gateway.

  • One API key for every model. No juggling provider accounts.
  • Three native protocols — OpenAI (/v1), Anthropic (/anthropic), and Gemini (/gemini). Your existing SDK works without code changes.
  • Auto failover — if the primary provider is down, requests route to a backup automatically. No downtime, no manual intervention.
  • Transparent pricing — upstream cost multiplied by 1.05. No hidden fees, no markup surprises. Per-request cost is queryable via the Generation API.
  • Dashboard — monitor usage, manage API keys, view request logs, and track billing at app.chuizi.ai.

How It Works

Your App  →  api.chuizi.ai  →  OpenAI / Anthropic / Google / DeepSeek / ...
              (gateway)         (upstream providers)

You send requests to Chuizi.AI using your preferred protocol. The gateway authenticates your ck- key, routes the request to the right upstream provider, and streams the response back. Billing is calculated from the upstream provider's reported token usage.

Supported Protocols

ProtocolBase URLUse Case
OpenAI compatiblehttps://api.chuizi.ai/v1Most SDKs, LangChain, LlamaIndex, n8n
Anthropic nativehttps://api.chuizi.ai/anthropicClaude Code, Cline, Anthropic SDK
Gemini nativehttps://api.chuizi.ai/geminiGoogle AI SDK, Gemini CLI

All three protocols share the same authentication, billing, and rate limiting. A request made via /anthropic appears in the same dashboard and uses the same balance as one made via /v1.

Responses API: Models that only support the Responses API (GPT-5 Pro, GPT-5 Codex, etc.) are transparently supported via the /v1 path — the gateway automatically converts chat/completions requests to Responses API format and converts responses back. No extra configuration needed.

Next Steps