I created a free open-source web extension to run local AI models

Hey fellow developers! 👋 I’m excited to introduce Ollamazing, a browser extension that brings the power of local AI models directly into your browsing experience. Let me share why you might want to give it a try.

What is Ollamazing?

Ollamazing is a free, open-source browser extension that connects with Ollama to run AI models locally on your machine. Think of it as having ChatGPT-like (or even Deepseek for newer) capabilities, but with complete privacy and no subscription fees.

🌟 Key Features

1. 100% Free and Open Source
   • No hidden costs or subscription fees
   • Fully open-source codebase
   • Community-driven development
   • Transparent about how your data is handled
2. Local AI Processing
   • Thanks to Ollama, we can run AI models directly on your machine
   • Complete privacy - your data never leaves your computer
   • Works offline once models are downloaded
   • Support for various open-source models (llama3.3, gemma, phi4, qwen, mistral, codellama, etc.) and specially deepseek-r1 - the most popular open source model at current time.
3. Seamless Browser Integration
   • Chat with AI right from your browser sidebar
   • Text selection support for quick queries
   • Context-aware responses based on the current webpage
4. Developer-Friendly Features
   • Code completion and explanation
   • Documentation generation
   • Code review assistance
   • Bug fixing suggestions
   • Multiple programming language support
5. Easy Setup
   • Install Ollama on your machine or any remote server - don’t forget to set up the `OLLAMA_ORIGINS`
   • Download your preferred models
   • Install the Ollamazing browser extension
   • Start chatting and using utilities with AI!

🚀 Getting Started

# 1. Install Ollama
curl -fsSL https://ollama.com/install.sh | sh

# 2. Pull your first model (e.g., Deepseek R1 7 billion parameters)
ollama pull deepseek-r1:7b

Then simply install the extension from your browser’s extension store.

For more information about Ollama, please visit the official website.

Important: If you run Ollama on local machine, ensure to setup the OLLAMA_ORIGINS to allow the extension can connect to the server. For more details, read Ollama FAQ, set the OLLAMA_ORIGINS to * or chrome-extension://* or the domain you want to allow.

💡 Use Cases

• Documentation generation
• Page content summary
• Document, Code review assistance

🔒 Privacy First

Unlike cloud-based AI assistants, Ollamazing:

• Keeps your data on your machine
• Doesn’t require an internet connection for inference
• Gives you full control over which model to use
• Allows you to audit the code and know exactly what’s happening with your data

🛠️ Technical Stack

• Use framework WXT to build the extension
• Built with React and TypeScript
• Uses Valtio for state management
• Implements TanStack Query for efficient data fetching
• Follows modern web extension best practices
• Utilizes Shadcn/UI for a clean, modern interface
• Use i18n for multi-language support

🤝 Contributing

We welcome contributions! Whether it’s:

• Adding new features
• Improving documentation
• Reporting bugs
• Suggesting enhancements

Check out our GitHub repository https://github.com/buiducnhat/ollamazing to get started!

🔮 Future Plans

We’re working on:

• Enhanced context awareness
• Custom model fine-tuning support
• Improve UI/UX
• Improved performance optimizations
• Additional browser support

🚀 Try It Today!

Ready to experience local AI in your browser? Get started with Ollamazing:

• Chrome web store: https://chromewebstore.google.com/detail/ollamazing/bfndpdpimcehljfgjdacbpapgbkecahi
• GitHub repository: https://github.com/buiducnhat/ollamazing
• Product Hunt: https://www.producthunt.com/posts/ollamazing

Let me know in the comments if you have any questions or feedback! Have you tried running AI models locally before? What features would you like to see in Ollamazing?