Get your Qwen API key in five minutes

What a Qwen API key is - and when you need one

A quick orientation so you don't set up something you don't actually need.

A Qwen API key is a secret credential - a string beginning with sk- - that authenticates your code when it calls Qwen's models programmatically. It is issued through Alibaba Cloud Model Studio, the developer platform for everything Qwen, whose underlying API surface is called DashScope (you'll see both names in the docs; they mean the same service). Every request your application sends includes this key so the platform knows who you are and what to bill.

The important thing to know first: you only need a key if you're a developer building Qwen into your own software. If you just want to chat with Qwen - ask questions, write, analyse documents and images - you don't need a key at all. Open the free Qwen Chat in your browser and start typing. The key is exclusively for the API: calling the models from a script, an app, or a production backend. If that's you, the rest of this page is the complete path from zero to your first working request.

Get your API key, step by step

Six steps, about five minutes. Have your email and phone handy for verification.

# macOS / Linux - current session only
export DASHSCOPE_API_KEY="sk-your-key-here"

# Make it permanent - add to ~/.bashrc or ~/.zshrc
echo 'export DASHSCOPE_API_KEY="sk-your-key-here"' >> ~/.bashrc

# Windows PowerShell
$env:DASHSCOPE_API_KEY = "sk-your-key-here"

Is the Qwen API key free?

Yes - and the free allowance is generous enough to genuinely prototype with, not just kick the tyres.

New Alibaba Cloud Model Studio accounts in the International (Singapore) region receive 1 million input tokens and 1 million output tokens free, valid for 90 days after you activate Model Studio. That's enough to thoroughly prototype an application, benchmark several models against each other, and prove out an architecture before paying anything. A few details worth knowing so it doesn't surprise you later:

• It applies only to the International (Singapore) endpoint - the US (Virginia) and mainland China regions have no free quota.
• The 1M + 1M is a combined cap across most Qwen models, not per-model - spend it however you like across Max, Plus, Turbo, Coder, VL, and the rest.
• It expires after 90 days whether or not you use it, so don't sit on it.
• Beyond the free quota, billing kicks in automatically using the payment method you added at signup - no surprise overage, since you set up billing during onboarding.
• Reasoning models emit far more output tokens than you might expect, which burns the output half faster - keep an eye on usage in the console.

Your first call with the key

The API is OpenAI-compatible, so your key drops straight into the standard SDK - change three things and you're live.

Install the official openai package, point its base URL at the Qwen endpoint, and pass your key. The three things you change versus a normal OpenAI setup are the base URL, the API key, and the model name - everything else (streaming, function calling, JSON mode) works identically.

Install the SDK

pip install openai

Make a request

import os
from openai import OpenAI

client = OpenAI(
    api_key=os.getenv("DASHSCOPE_API_KEY"),   # your sk- key
    base_url="https://dashscope-intl.aliyuncs.com/compatible-mode/v1",
)

completion = client.chat.completions.create(
    model="qwen-plus",
    messages=[
        {"role": "system", "content": "You are a helpful assistant."},
        {"role": "user", "content": "Say hello and confirm my key works."}
    ]
)

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

If that prints a reply, your key is live and authenticated. From here every standard feature is available through the same client - streaming responses, tool/function calling, JSON mode, multi-turn conversations, and image inputs on the vision models.

curl https://dashscope-intl.aliyuncs.com/compatible-mode/v1/chat/completions \
  -H "Authorization: Bearer $DASHSCOPE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"qwen-plus","messages":[{"role":"user","content":"Hi"}]}'

Your key is tied to a region

Model Studio runs in four regions, each with its own endpoint and its own keys - and they are not interchangeable.

This is the single most common source of confusion, so it's worth getting right up front. A key created on the Singapore endpoint will fail authentication on the Beijing or US endpoints, and vice versa. Pick the region closest to your users - or whichever your compliance team approves - and use that base URL everywhere. If you ever get a 401 error, a mismatched base URL is more likely the cause than a bad key.

Protecting your API key

A leaked key is a leaked credit card - anyone who has it can spend on your account. A few habits keep that from happening.

Your key is a secret that authorizes billable usage, so treat it with the same care as a password. The good news is that safe handling is mostly a matter of a few simple, durable habits - set them once and you rarely have to think about it again.

• Never hardcode it in source files - read it from an environment variable or secrets manager instead.
• Never commit it to Git, especially a public repo. Add .env to your .gitignore before your first commit.
• Never expose it in client-side code - browser or mobile apps should call your own backend, which holds the key, never ship it to the device.
• Use separate keys for production, staging, and each application, so you can revoke one without breaking the others.
• Rotate and revoke from the API Keys page if a key is ever exposed - revoke the old one and issue a new one immediately.
• Monitor usage in the Model Studio console so an unexpected spike (a sign of a leak) gets caught early.

Frequently asked questions