Integrate with Qwen Code

June 29, 2026 · View on GitHub

Qwen Code is an open-source AI agent that lives in your terminal, built by the Qwen team at Alibaba Group.

Prerequisites

Both Pay-as-you-go API and Token Plan are supported with Qwen Code. You need to obtain the corresponding credentials before configuration.

Usage Mode	Description	How to Get Credentials
Pay-as-you-go	Billed by actual usage, suitable for light use	Go to API Keys and create an API Key
Token Plan	Fixed subscription with quota-based access	After subscribing, go to Subscription Management to get your dedicated Base URL and API Key

npm install -g @qwen-code/qwen-code@latest

Or via the quick-install script:

curl -fsSL https://qwen-code-assets.oss-cn-hangzhou.aliyuncs.com/installation/install-qwen-standalone.sh | bash

Or Homebrew:

brew install qwen-code

qwen --version

Launch Qwen Code:

qwen

Inside the interactive session, run /auth:

/auth

Select Custom Provider and follow the 6-step wizard:

Step	Field	Value
1	Protocol	OpenAI Compatible
2	Base URL	`https://api.xiaomimimo.com/v1` (Pay-as-you-go) or `https://token-plan-{region}.xiaomimimo.com/v1` (Token Plan)
3	API Key	Your MiMo API Key
4	Model ID	`mimo-v2.5-pro`
5	Advanced Config	Enable thinking: on; Enable modality: off; Context window: `1048576`
6	Provider Name	`MiMo` (or any name you prefer)

Run the /model command and select mimo-v2.5-pro:

/model

You're ready to go. Qwen Code will now use MiMo for coding.

Tip: Use /model to switch between models at any time. Use /auth again if you need to reconfigure.