Updated: July 29, 2026 · 8 min read

How to Connect Cursor, Claude Code & Any Tool to Chinese LLMs

Chinese LLMs like DeepSeek V4, Qwen3, and GLM-4 offer exceptional performance at a fraction of the cost of Western alternatives. But connecting them to your favorite development tools isn't always straightforward.

This guide shows you how to use a single OpenAI-compatible endpoint to access all major Chinese models from Cursor, Claude Code, Open WebUI, LangChain, CrewAI, and more.

Why Chinese LLMs?

Model	Input Price	vs GPT-4o	Strength
DeepSeek V4	$0.28/M tok	94% cheaper	Coding, math, reasoning
Qwen3.6 Flash	$0.07/M tok	98% cheaper	Fast, budget-friendly
GLM-4 Flash	$0.07/M tok	98% cheaper	Multilingual
Kimi K3	$0.14/M tok	97% cheaper	Long context (128K)

The Solution: AI Token Hub

AI Token Hub provides a single OpenAI-compatible endpoint that routes to all major Chinese LLMs. One API key, one base URL, all models.

Base URL: https://eaf9553505eeb8f5-115-190-107-107.serveousercontent.com/v1

Tool-by-Tool Setup

1. Cursor IDE

Settings → Models → Add Model:
  Model name: deepseek-v4
  Base URL: https://eaf9553505eeb8f5-115-190-107-107.serveousercontent.com/v1
  API key: YOUR_KEY

2. Claude Code

export ANTHROPIC_BASE_URL="https://eaf9553505eeb8f5-115-190-107-107.serveousercontent.com/v1"
export ANTHROPIC_AUTH_TOKEN="YOUR_KEY"
export ANTHROPIC_API_KEY=""

3. Open WebUI

Settings → Connections → OpenAI API:
  URL: https://eaf9553505eeb8f5-115-190-107-107.serveousercontent.com/v1
  Key: YOUR_KEY

4. LangChain (Python)

from openai import OpenAI
client = OpenAI(
    base_url="https://eaf9553505eeb8f5-115-190-107-107.serveousercontent.com/v1",
    api_key="YOUR_KEY"
)
response = client.chat.completions.create(
    model="deepseek-v4",
    messages=[{"role": "user", "content": "Hello!"}]
)

5. CrewAI

import os
os.environ["OPENAI_API_BASE"] = "https://eaf9553505eeb8f5-115-190-107-107.serveousercontent.com/v1"
os.environ["OPENAI_API_KEY"] = "YOUR_KEY"

from crewai import LLM
llm = LLM(model="deepseek-v4")

6. cURL (Quick Test)

curl https://eaf9553505eeb8f5-115-190-107-107.serveousercontent.com/v1/chat/completions \
  -H "Authorization: Bearer YOUR_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"deepseek-v4","messages":[{"role":"user","content":"Hello!"}]}'

Available Models

Model	Best For	Context
deepseek-v4	Flagship coding & reasoning	128K
deepseek-v3	Code generation	128K
deepseek-r1	Chain-of-thought reasoning	128K
qwen3.6-flash	Fast, cheap tasks	128K
qwen3-flash	Budget batch processing	128K
qwen-plus	Balanced performance	128K
qwen-max	Complex tasks	128K
glm-4-flash	Multilingual	128K
glm-4-air	General purpose	128K
kimi-k3	Long documents	128K

Get started free — No credit card required

Get Your API Key →

FAQ

Is this official DeepSeek/Qwen API?

No. AI Token Hub is a third-party gateway that proxies requests to Chinese model providers. It's useful for overseas developers who can't easily register for Chinese platforms directly.

What about data privacy?

See our Privacy Policy. We don't store conversation content. Requests are proxied in real-time.

Can I use this with any OpenAI-compatible tool?

Yes. Any tool that accepts a custom base_url and api_key will work. Just point it to our endpoint.

← Back to Blog | Developer Guides | GitHub