One API to Rule All Chinese LLMs: DeepSeek, Qwen, GLM, Kimi

If you're a developer working with Chinese LLMs, you know the pain: every model provider has its own API format, SDK, and authentication.

TokenHub solves this by providing a single OpenAI-compatible API endpoint that routes to all major Chinese LLMs.

What's included

• DeepSeek - Most cost-effective
• Qwen (Tongyi Qianwen) - 238 models via DashScope
• GLM (Zhipu) - Fast response, free Flash tier
• Kimi (Moonshot) - 128K context window

How it works

pip install openai

from openai import OpenAI
client = OpenAI(
    api_key="your_tokenhub_key",
    base_url="https://loyap-innovation.com/v1/"
)

# Switch models by changing the model name only
response = client.chat.completions.create(
    model="deepseek-chat",  # or qwen-plus, glm-4-flash, moonshot-v1-128k
    messages=[{"role": "user", "content": "Hello"}]
)

Key Benefits

• RMB pricing - Pay in Chinese Yuan directly
• No VPN - Works from mainland China
• Unified billing - One balance for all models
• Auto failover - If one channel goes down, traffic routes to another

Try it live: https://loyap-innovation.com/chat/

Get your API key: https://loyap-innovation.com/token/

GitHub: https://github.com/komine1728/tokenhub-chat-demo

TokenHub - AI API Gateway for Chinese LLMs