KI
KI Katapult

Bye Lovable.
Hallo Qwen Code.

Your step-by-step wizard to replace Lovable with a free, open-source AI coding setup in VS Code — no developer experience needed.

Step 1 of 70%
⚡️ Let's start

What's your operating system?

We'll tailor every download link and terminal command to your computer.

🪟WindowsWindows 10 / 11
🍎macOSVentura, Sonoma, Sequoia
Step 1

Install Visual Studio Code

VS Code is the free code editor where your AI assistant will live. Think of it as your new Lovable workspace — but on your computer.

  • Open the VS Code download page:
    ↗️ code.visualstudio.com/download
  • Click the big blue "Windows" button to download the installer. Run the .exe file and click "Next" through the wizard. Check ✅ "Add to PATH" when asked.Click the big blue "macOS" button. Unzip the download, then drag Visual Studio Code.app into your Applications folder.
  • Launch VS Code. You should see a Welcome tab. That's it — VS Code is ready!
💡 Already have VS Code?
Skip this step and continue. Just make sure it's a recent version (1.85+).
Step 2

Install Node.js

Node.js is the engine that runs Qwen Code. You need version 20 or higher. This is a one-time setup — 3 minutes tops.

  • Click the big green "LTS" (Long Term Support) button.
  • Run the downloaded .msi installer. Click "Next" through everything — the defaults are fine. ✅ Make sure "Add to PATH" is checked.Run the downloaded .pkg installer. Follow the prompts — all defaults are fine.
  • Verify it worked. Open a terminal and type:
    node --version
    You should see something like v22.x.x. As long as it's 20 or higher, you're good.
How to open a terminal:
Windows: Press Win + R, type cmd, press Enter.Mac: Press Cmd + Space, type Terminal, press Enter.
Step 3

Install Qwen Code (the AI brain)

This is the free, open-source AI coding agent that replaces Lovable's AI. It uses Alibaba's Qwen3-Coder model — one of the best coding AIs available, and it's 100% free with 1,000 requests per day.

  • Open your terminal and paste this command:
    npm install -g @qwen-code/qwen-code@latest
    Press Enter and wait about 30 seconds.
  • Verify it installed:
    qwen --version
  • Now sign in (free). Run:
    qwen
    Choose "Qwen OAuth" — your browser will open. Create a free account at qwen.ai, authorize, and you're connected.
What you just got: An AI coding agent that can read, write, and edit entire projects. It understands your codebase, creates files, runs commands, and works like a developer.
💡 Tip for power users
Qwen OAuth gives you 1,000 free requests/day. For heavier use, you can also use an OpenRouter API key with the model qwen/qwen3-coder.
Step 4

Connect Qwen Code to VS Code

This extension lets Qwen Code see your files, show changes visually, and work right inside VS Code — like Lovable's editor, but better.

  • Open VS Code. Click the Extensions icon in the left sidebar.
    Or press Ctrl + Shift + XOr press Cmd + Shift + X
  • In the search box, type:
    "Qwen Code Companion"
    (publisher: qwenlm)
  • Click Install. Wait a few seconds — done!
  • Open VS Code's built-in terminal: Ctrl + `Cmd + `
    Type qwen and press Enter. The extension will show "Connected to VS Code".
What this gives you: When Qwen Code edits files, you'll see the changes highlighted in VS Code's diff viewer — just like tracked changes in a document.
Step 5

Export your Lovable project

Time to get your code out of Lovable and onto your computer. We'll use GitHub as the bridge. If you're starting fresh, skip to the next step.

Part A — Connect Lovable to GitHub

  • If you don't have a GitHub account, create one (free) at ↗️ github.com/signup
  • Open your project in Lovable. Look for the GitHub icon in the top toolbar.
  • Click "Connect to GitHub" and authorize Lovable.
  • Click "Create Repository". Wait for the green checkmark ✅.

Part B — Download the code to your computer

  • Install Git (if you don't have it):
    ↗️ git-scm.com/download/win — run the installer with all defaults.Open Terminal and type git --version. macOS will prompt you to install it automatically.
  • Go to your GitHub repo page. Click the green "<> Code" button and copy the HTTPS URL.
  • Open your terminal and run:
    # Replace with YOUR repository URL: git clone https://github.com/YOUR-USERNAME/YOUR-PROJECT.git
  • Navigate into the project:
    cd YOUR-PROJECT
  • Install dependencies:
    npm install
About environment variables: Lovable doesn't export API keys for security reasons. If your project uses Supabase or other services, you'll need to create a .env.local file and add the keys there.
💡 Alternative: ZIP download
No Git? On your GitHub repo page, click the green "Code" button → "Download ZIP".
Step 6

Open your project with Qwen Code

This is where the magic happens. Let's fire up your project and start talking to your AI coding assistant.

  • Open VS Code. Go to File → Open Folder and select your project folder.
  • Open the terminal inside VS Code:
    Ctrl + `Cmd + `
  • Start Qwen Code:
    qwen
  • Now talk to it like you would to Lovable! Try:
    # Understand the project: What does this project do? Explain the codebase structure. # Make changes (like in Lovable's chat): Add a dark mode toggle to the header. # Run it locally: Start the development server so I can preview the app.
  • To preview your app locally:
    npm run dev
    Then open localhost:5173 in your browser.
You're now doing what Lovable does — but for free, on your machine, with full control. Qwen Code reads your files, writes code, creates components, fixes bugs, and can even run terminal commands for you.
🚀
You're all set!
Your Lovable replacement is ready. Here's what you can do now:
💬
Chat to build
Just describe what you want in natural language — like Lovable, but in your terminal.
👁
See changes live
The VS Code extension shows diffs for every change Qwen makes.
💰
Zero cost
1,000 free API calls per day. No subscription. No vendor lock-in.
🔒
You own everything
Code stays on your machine. Deploy anywhere you want.

Quick Reference — Daily Workflow

1. Open your project folder in VS Code
2. Open the terminal (Ctrl+`Cmd+`) and type qwen
3. Ask it to build, fix, or explain anything
4. Preview with npm run dev
5. Commit your changes with git add . && git commit -m "your message" && git push