Skip to main content
This guide walks you through setting up BrowserQwen, including deploying the local database service and installing the Chrome extension.

​
Prerequisites

  • Chrome browser
  • Python 3.8 or higher
  • Qwen-Agent installed
  • API access to DashScope or a self-hosted model service

​
Step 1: Deploy Local Database Service

The local database service manages your browsing history and conversation history. This service must be running on the same machine where you use Chrome.

​
Using DashScope Model Service

If you’re using Alibaba Cloud’s DashScope:
1

Set up API Key

Obtain your DashScope API key from DashScope Console
2

Start the Database Service

Run the following command:
python run_server.py --llm qwen-max --model_server dashscope --workstation_port 7864 --api_key YOUR_DASHSCOPE_API_KEY
Replace YOUR_DASHSCOPE_API_KEY with your actual API key.
3

Choose Model

The --llm parameter accepts various models in increasing order of resource consumption:
  • qwen1.5-7b-chat / qwen1.5-14b-chat / qwen1.5-72b-chat
  • qwen-turbo
  • qwen-plus
  • qwen-max (recommended)

​
Using Self-Hosted Model Service

If you’re using your own model service (e.g., vLLM, Ollama):
1

Deploy Your Model

Deploy your model using vLLM or another OpenAI-compatible service. For example, with vLLM:
python -m vllm.entrypoints.openai.api_server \
  --model Qwen/Qwen1.5-72B-Chat \
  --host localhost \
  --port 8000
2

Start the Database Service

Point to your model service:
python run_server.py \
  --llm Qwen1.5-72B-Chat \
  --model_server http://localhost:8000/v1 \
  --workstation_port 7864 \
  --api_key EMPTY
Replace {MODEL}, {API_BASE}, and {API_KEY} with your actual values.

​
Verify Service is Running

Once the service is running, you can access the Workstation interface at: 
http://127.0.0.1:7864/
This gives you access to:
  • Editor mode for long-form content creation
  • Chat mode for multi-document QA

​
Step 2: Install Browser Extension

Now install the BrowserQwen Chrome extension:
1

Open Chrome Extensions Page

Navigate to Chrome’s extensions page:
chrome://extensions/
Or use the menu: More tools → Extensions
2

Enable Developer Mode

Toggle Developer mode in the top right corner to ON
3

Load Extension

Click Load unpacked and select the browser_qwen directory from your Qwen-Agent installation
The browser_qwen directory is located in the root of the Qwen-Agent repository.
4

Pin Extension

Click the extensions icon (puzzle piece) in Chrome’s toolbar and pin BrowserQwen for easy access
After installing the extension, refresh any open web pages for the extension to take effect on those pages.

​
Using BrowserQwen

Once both the service and extension are installed:
1

Add Page to Reading List

On any webpage or PDF, click the Add to Qwen’s Reading List button that appears on the screen. This authorizes Qwen to analyze the page in the background.
2

Start Chatting

Click the BrowserQwen icon in your browser’s toolbar to open the chat interface and start asking questions about the current page or your browsing history.

​
Troubleshooting

  • Make sure Developer mode is enabled
  • Verify you selected the correct browser_qwen directory
  • Try restarting Chrome
  • Ensure the run_server.py process is still running
  • Check that port 7864 is not blocked by firewall
  • Verify the service is accessible at http://127.0.0.1:7864/
  • Refresh the page after installing the extension
  • Some pages (like chrome:// URLs) don’t allow extensions
  • Check browser console for errors

​
Next Steps

Explore Features

Learn about all the features and capabilities of BrowserQwen

Build docs developers (and LLMs) love

Get started for freeTalk to us