ChinaWHAPI Team

Posted on May 12

llm

#ai #llm #api #machinelearning

Many developers outside China want to test DeepSeek, but setup can be difficult due to different APIs, Chinese documentation and access issues.

ChinaWH API makes it easier by providing an OpenAI-compatible endpoint for Chinese LLMs including DeepSeek, Qwen and GLM.

Why Global Developers Need Chinese LLM API

Chinese large language models like DeepSeek, Qwen, and GLM have become powerful options for AI applications, especially for:

Chinese language tasks (writing, translation, customer support)
Cost-sensitive AI workflows
AI agents requiring strong reasoning capabilities
Chatbots serving Chinese-speaking users

However, accessing these models from outside China can be challenging:

Different API formats across providers
Chinese-only documentation
Complex authentication flows
Inconsistent integration patterns

The Solution: OpenAI-Compatible API Gateway

If your application already supports OpenAI-compatible APIs, you can access Chinese LLMs with minimal code changes. The key is using an OpenAI-compatible gateway that maintains the /v1/chat/completions format.

Introducing ChinaWH API

ChinaWH API provides a unified, OpenAI-compatible API gateway for Chinese LLMs. With one API key, you can access:

DeepSeek - Strong reasoning and coding capabilities
Qwen (Alibaba) - Excellent Chinese language understanding
GLM (Zhipu) - Balanced performance across tasks
Moonshot - Long context support
ERNIE (Baidu) - Enterprise-grade Chinese NLP
Doubao (ByteDance) - Consumer-focused AI
MiniMax - Multimodal capabilities

Quick Start Guide

Step 1: Get Your API Key

Step 2: Change Your Base URL

Instead of https://api.openai.com/v1, use:

https://api.chinawhapi.com/v1

Step 3: Update Model Name

Replace gpt-3.5-turbo with the Chinese model you want to use:

deepseek-chat for DeepSeek
qwen-plus for Qwen
glm-4 for GLM
And more...

Code Examples

Python Example

from openai import OpenAI

client = OpenAI(
    api_key="your_chinawhapi_key",
    base_url="https://api.chinawhapi.com/v1"
)

response = client.chat.completions.create(
    model="deepseek-chat",
    messages=[
        {"role": "user", "content": "Hello, how are you?"}
    ]
)

print(response.choices[0].message.content)

Node.js Example

const { OpenAI } = require('openai');

const client = new OpenAI({
  apiKey: 'your_chinawhapi_key',
  baseURL: 'https://api.chinawhapi.com/v1'
});

const response = await client.chat.completions.create({
  model: 'deepseek-chat',
  messages: [{ role: 'user', content: 'Hello, how are you?' }]
});

console.log(response.choices[0].message.content);

DEV Community

llm