How to Use Alibaba's Qwen API Outside China

Alibaba's Qwen (Tongyi Qianwen) family of models has become one of the strongest contenders in the AI landscape, especially for multilingual tasks, long-context understanding, and instruction following. In many benchmarks, Qwen-Max rivals GPT-4o and Claude 3.5 Sonnet.

The challenge for international developers? Qwen's official API is served through Alibaba Cloud (Aliyun), which typically requires Chinese identity verification and payment through Chinese channels. This effectively locks out most developers outside China.

Available Qwen Models via NovAI

Quick Start: Access Qwen in 3 Steps

Step 1: Sign up at aiapi-pro.com with your email. No phone number or identity verification needed.

Step 2: Get your API key from the dashboard (starts with nvai-).

Step 3: Make your first API call:

from openai import OpenAI

client = OpenAI(
    api_key="nvai-your-api-key",
    base_url="https://aiapi-pro.com/v1"
)

# Try Qwen-Max for strongest performance
response = client.chat.completions.create(
    model="qwen-max",
    messages=[
        {"role": "system", "content": "You are a helpful assistant."},
        {"role": "user", "content": "Compare React and Vue for a new project"}
    ],
    temperature=0.7
)

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

Why Qwen? Strengths by Use Case

Multilingual excellence: Qwen was trained with a strong emphasis on Chinese and English, but it also performs well in Japanese, Korean, French, German, Spanish, and more. If you're building multilingual products, Qwen is arguably the best value.

Long context: Qwen-Max supports up to 32K context window, and Qwen-Turbo handles 128K tokens. This makes it excellent for document analysis, code review, and conversation summarization.

Instruction following: Qwen models excel at structured output, JSON generation, and following complex multi-step instructions — critical for building reliable AI agents and pipelines.

Qwen vs Other Models: When to Use What

The beauty of NovAI is that all these models are accessible through the same API key and endpoint. You can route different tasks to different models based on their strengths.

Streaming Example with Qwen-Turbo

import httpx, json

url = "https://aiapi-pro.com/v1/chat/completions"
headers = {
    "Authorization": "Bearer nvai-your-api-key",
    "Content-Type": "application/json"
}
payload = {
    "model": "qwen-turbo",
    "messages": [{"role": "user", "content": "Write a haiku about programming"}],
    "stream": True
}

with httpx.stream("POST", url, json=payload, headers=headers) as r:
    for line in r.iter_lines():
        if line.startswith("data: ") and "[DONE]" not in line:
            chunk = json.loads(line[6:])
            delta = chunk["choices"][0]["delta"].get("content", "")
            print(delta, end="", flush=True)