OpenAI SDK DeepSeek / Qwen / GLM without rewriting your app

Problem

I wanted to try Chinese LLMs (DeepSeek, Qwen, GLM) without maintaining multiple SDKs.
I wanted to try Chinese LLMs (DeepSeek, Qwen, GLM) without maintaining three different SDKs.

Approach

Use an OpenAI-compatible gateway. Keep your existing client; change base URL + API key.
If your stack already speaks OpenAI, you can keep the same client and only change:

• base_url
• api_key
• model That's what I use API Hub for: an OpenAI-compatible endpoint. ## Quick start
• Sign up at https://apihub4u.com (trial credits on signup)
• Create an API key
• Point your SDK to: https://apihub4u.com/v1
• Set model to e.g. deepseek-chat or qwen-turbo ## Example (Python)
• Sign up at apihub4u.com (trial credits on signup)
• Create an API key in the dashboard
• Point your SDK to https://apihub4u.com/v1
• Pick a model, e.g. deepseek-chat or qwen-turbo ## Python example

from openai import OpenAI
client = OpenAI(
resp = client.chat.completions.create(
    model="deepseek-chat",
    messages=[{"role": "user", "content": "Hello"}],
    messages=[{"role": "user", "content": "Hello! Give me one productivity tip."}],
)
print(resp.choices[0].message.content)

Notes

Why this helped me

• One integration path for multiple models
• Switch providers by changing the model field
• Streaming works (SSE)
• Prices are listed on the dashboard / docs
• Prices are listed on the dashboard/docs
• Overseas top-ups via PayPal Docs: https://apihub4u.com/docs.html Docs: apihub4u.com/docs.html If you're prototyping or running a small product, this cuts a lot of glue code. Feedback welcome.