Beta · $1 free credit on signup

One API Key. Every Model.

Yueyou Tech AI TOKEN aggregates Zhipu GLM · MiniMax · DeepSeek behind a single standard OpenAI-compatible endpoint. Copy-paste ready, with one-click setup for Cherry Studio / LobeChat / OpenCat.

👨‍💻 Developers: route Claude Code / Codex CLI through CC Switch to Yueyou TOKEN and code with GLM / DeepSeek.

Sign up · Get $1 View Setup Guide
8top models
$1free signup credit
OpenAI-compatible API
$1/$10top-up tiers
Why us

Why Yueyou TOKEN

🎁

Zero-friction trial

Get $1 credit on signup — no credit card required. Call any model's API immediately to validate quality.

🔌

Standard API

Fully OpenAI API compatible. Mainstream clients and code frameworks work without any modification.

🌐

Model aggregation

Zhipu GLM, MiniMax, DeepSeek in one place. Call flexibly by model — no need to open accounts everywhere.

Models

Supported Models

Click any model name to copy it into the Model field of your client

🧠 Zhipu GLM

glm-5glm-4.7glm-4.7-flash

💡 MiniMax

MiniMax-M3MiniMax-M2.7MiniMax-M2.7-highspeed

🔥 DeepSeek

deepseek-v4-prodeepseek-v4-flash
Quick Start

Quick Start · 4 Steps

Sign up & claim $1 credit

Open /register to finish signup; the system auto-grants $1 free trial credit (no credit card needed).

Create your API Key

After login, go to left menu Token / API Key → click "Create Token" → copy the generated sk-... key (shown only once, keep it safe).

Fill config in your client

Open any client below, add a "Custom / OpenAI-compatible" provider, and fill in the endpoint URL and your key. See each client card for details.

Send your first message

Pick a model (e.g. glm-4.7) and start chatting, or verify connectivity with the curl command below.

One-click Access

Client Setup

All three clients support OpenAI-compatible mode with identical config fields. Developers: scroll down to the CC Switch developer section

Cherry Studio

Desktop · Windows / macOS / Linux
Endpointhttps://www.yueyoukejitoken.com/v1
API KeyPaste your real key from Step 2 (sk-...)
Modelglm-4.7 (any of the 8)
Steps:
  1. Settings → Model Providers → add "OpenAI"
  2. API Key = your sk-..., Base URL = endpoint above
  3. Model list → "Manage" → add the model name you want

LobeChat

Web / Desktop · Open Source
Endpointhttps://www.yueyoukejitoken.com/v1
API KeyPaste your real key from Step 2 (sk-...)
Modelglm-4.7 (any of the 8)
Steps:
  1. Session Settings → Model → choose "OpenAI"
  2. Fill custom Endpoint and API Key
  3. Save, then select our model from the dropdown

OpenCat

iOS / macOS client
Endpointhttps://www.yueyoukejitoken.com/v1
API KeyPaste your real key from Step 2 (sk-...)
Modelglm-4.7 (any of the 8)
Steps:
  1. Settings → API Provider → add "OpenAI-compatible"
  2. Base URL = endpoint above, Key = your sk-...
  3. Refresh model list, pick our model to chat

CC Switch ⚙️ Developer Mode

Windows / macOS / Linux · requires Claude Code / Codex CLI

CC Switch ≠ Cherry Studio. It is a config manager for AI coding CLIs — it manages base URL, API key and model switching for Claude Code / Codex / Gemini CLI / OpenCode. It does not send chat requests itself.

Through CC Switch's built-in local proxy (port 15721), it converts the OpenAI protocol to Anthropic for Claude Code — so even though Yueyou TOKEN only exposes an OpenAI-compatible endpoint, Claude Code works.

⚠️ Only for developers already using Claude Code / Codex CLI. Casual chat users should use the three clients above.
AppClaude (in Advanced, pick OpenAI Chat Completions format)
Base URLhttps://www.yueyoukejitoken.com/v1
API KeyPaste your real key from Step 2 (sk-...)
Auth fieldANTHROPIC_AUTH_TOKEN (default)
Steps:
  1. Install Claude Code: npm i -g @anthropic-ai/claude-code
  2. Download CC Switch desktop app (GitHub Releases: .msi / .dmg)
  3. CC Switch → Claude tab → ➕ top-right → fill Base URL / API Key
  4. Advanced → API format = OpenAI Chat Completions
  5. Settings → Proxy → enable the master switch (default 127.0.0.1:15721)
  6. Run claude to verify — protocol is converted by the local proxy
🧩 Model Mapping (Claude → Yueyou TOKEN)
Sonnet / main→deepseek-v4-pro
Opus / heavy→glm-5
Haiku / light→glm-4.7-flash
Thinking→deepseek-v4-pro
Test

Verify Connectivity

Replace YOUR_API_KEY with your real key, then run in terminal:

curl https://www.yueyoukejitoken.com/v1/chat/completions \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"glm-4.7","messages":[{"role":"user","content":"hello"}],"stream":false}'
FAQ

Frequently Asked Questions

How is the $1 credit charged?

It is a trial credit issued by the platform. Model calls consume it by token usage and do not deduct the admin balance; the real cost is the upstream model channel, borne by the platform as acquisition cost.

Which models are supported?

8 models today: Zhipu glm-5 / glm-4.7 / glm-4.7-flash; MiniMax M3 / M2.7 / M2.7-highspeed; DeepSeek v4-pro / v4-flash.

What API standard?

Fully OpenAI-compatible (e.g. /v1/chat/completions). Mainstream SDKs (openai, langchain) and clients work directly.

How to top up?

During Beta, only $1 / $10 tiers are available. More tiers open after official launch.

Is CC Switch the same as Cherry Studio?

No. Cherry Studio is an AI chat client — open and chat. CC Switch is a config manager for AI coding CLIs — it manages base URL, API key and model switching for Claude Code / Codex / Gemini CLI. Different audiences: casual users use Cherry Studio / LobeChat; developers configuring Claude Code use CC Switch.

📘 Setup Guide