Overview
Qwen Code supports OpenAI-compatible model providers through ~/.qwen/settings.json. Add Vtrix as an OpenAI-compatible provider, set the Vtrix base URL, and select a Vtrix model.
Prerequisites
- Node.js 22 or higher when installing with npm
- Qwen Code
- A Vtrix API key, created in Settings
You can also use Vtrix Doctor to check the local Base URL, API key, and model settings before editing files. Supported tools, including OpenClaw and Hermes, can be configured after confirmation.
Configuration Steps
Install Qwen Code
• Using the install script
curl -fsSL https://qwen-code-assets.oss-cn-hangzhou.aliyuncs.com/installation/install-qwen-standalone.sh | bash• Or using npm
npm install -g @qwen-code/qwen-code@latestAfter installation, run:
qwen --versionSet your Vtrix API key
Add the following to ~/.bashrc or ~/.zshrc:
export VTRIX_API_KEY="YOUR_VTRIX_API_KEY"After updating the file, restart your terminal or run:
source ~/.zshrcConfigure the Vtrix provider
Create ~/.qwen/settings.json:
mkdir -p ~/.qwen
vim ~/.qwen/settings.jsonAdd the following configuration:
{
"modelProviders": {
"openai": {
"protocol": "openai",
"models": [
{
"id": "qwen3.7-max",
"name": "Qwen3.7 Max via Vtrix",
"baseUrl": "https://cloud.vtrix.ai/llm",
"description": "Qwen3.7 Max routed through the Vtrix OpenAI-compatible endpoint",
"envKey": "VTRIX_API_KEY"
}
]
}
},
"security": {
"auth": {
"selectedType": "openai"
}
},
"model": {
"name": "qwen3.7-max"
}
}Use https://cloud.vtrix.ai/llm as the OpenAI-compatible Vtrix base URL. Qwen Code reads the actual API key from the VTRIX_API_KEY environment variable named by envKey.
Start Qwen Code
Go to your project directory and start Qwen Code:
cd /path/to/your/project
qwenFor a one-off check, use headless mode:
qwen -p "Summarize this project in one sentence"If Qwen Code responds normally, requests are routed to Vtrix through the configured base URL. You can also check usage or billing in Vtrix Settings.
Troubleshooting
Error: 401 Unauthorized
Cause: VTRIX_API_KEY is missing, incorrect, or not loaded in the current terminal session.
Fix:
echo $VTRIX_API_KEYIf the value is empty, check your shell config and restart the terminal.
Qwen Code starts with a different provider
Cause: security.auth.selectedType is not set to openai, or another settings file overrides the user settings.
Fix: Confirm ~/.qwen/settings.json contains:
{
"security": {
"auth": {
"selectedType": "openai"
}
}
} Error: model not found
Cause: model.name or the model id does not match a Vtrix model ID.
Fix: Open the Models page, copy the correct model ID, and update both model.name and the matching modelProviders.openai.models[].id.
Other configuration issues
- Confirm
protocolisopenai - Confirm
baseUrlishttps://cloud.vtrix.ai/llm - Confirm
envKeyisVTRIX_API_KEY - Restart Qwen Code after editing
~/.qwen/settings.json