A Next.js application that demonstrates and tests browser-based AI APIs. Run advanced machine learning models directly in your browser with privacy, speed, and no server costs.
- On-Device AI: All AI processing happens locally in your browser
- Privacy-First: No data is sent to external servers
- Real-Time API Status: Check availability of browser AI APIs
- Interactive Playgrounds: Test each API with a user-friendly interface
| API | Description |
|---|---|
| Prompt API | General purpose language model for chat and text generation |
| Summarizer API | Summarize long texts into concise key points or summaries |
| Writer API | Generate new content based on context and prompts |
| Rewriter API | Rewrite and polish existing text with different tones |
| Language Detection | Detect the language of a given text |
- Node.js (v18 or higher recommended)
- A browser that supports Chrome Built-in AI APIs (e.g., Chrome Canary with experimental flags enabled)
-
Clone the repository:
git clone https://github.com/yogeshpaliyal/browser-ai.git cd browser-ai -
Install dependencies:
npm install
-
Run the development server:
npm run dev
-
Open http://localhost:3000 in your browser to see the application.
This application requires a browser with support for Chrome's Built-in AI APIs. Currently, these APIs are experimental and available in:
- Chrome Canary with the following flags enabled:
chrome://flags/#optimization-guide-on-device-modelchrome://flags/#prompt-api-for-gemini-nano
For the latest information on browser support and how to enable these features, refer to the Chrome Built-in AI documentation.
- Next.js - React framework
- React - UI library
- Tailwind CSS - Styling
- TypeScript - Type safety
MIT