Run Powerful LLMs in VS Code Using Google Colab GPU
If you’re a developer, there’s a high chance your favorite coding companion is Visual Studio Code. And why not?
VS Code gives you full control — from themes to extensions — and makes coding feel effortless.
But if you work in data science, machine learning, or LLM development, you’ve surely faced this issue:
- Running heavy models
- Processing large datasets
- Training deep learning pipelines
…and suddenly your laptop starts overheating, the fan roars like a jet engine, and everything slows down.
This is when we all turn to Google Colab — the magical cloud notebook environment that gives:
- Free GPUs
- Zero setup
- Easy collaboration
- Cloud storage
- Fast execution
It’s like having a supercomputer right inside your browser.
Get Yash Kavaiya’s stories in your inbox
Join Medium for free to get updates from this writer.
But switching back and forth between Colab and VS Code can be annoying.
What if both worlds came together?
1. Install the Extension
I’m using VS Code version 1.106.0.
The extension appear in the marketplace
2. Create a New Jupyter Notebook
- Create a file named demo.ipynb
- Click on the kernel selector
- Choose Colab
- Click Autoconnect
For the first-time setup, VS Code will ask you to sign in to your Google account.
Grant the permissions — and you’re done.
3. Enable GPU — Without Leaving VS Code
🎉 Final Thoughts
The new Google Colab extension for VS Code brings:
💻 VS Code convenience
⚡ Colab compute power
🤝 Seamless integration
🚀 Massive productivity boost
