|
|
---
|
|
|
title: FARA - Computer Use Agent
|
|
|
emoji: 🤖
|
|
|
colorFrom: blue
|
|
|
colorTo: purple
|
|
|
sdk: docker
|
|
|
pinned: false
|
|
|
license: mit
|
|
|
app_port: 7860
|
|
|
suggested_hardware: cpu-upgrade
|
|
|
tags:
|
|
|
- computer-use
|
|
|
- browser-automation
|
|
|
- ai-agent
|
|
|
- vision-language-model
|
|
|
---
|
|
|
|
|
|
# 🤖 FARA - Computer Use Agent Demo
|
|
|
|
|
|
FARA (Fara Agent for Real-world Automation) is an AI agent that can browse the web and complete tasks autonomously.
|
|
|
|
|
|
## Features
|
|
|
|
|
|
- 🌐 **Autonomous Web Navigation** - The agent can browse websites on its own
|
|
|
- 🔍 **Web Search** - Search for information across the web
|
|
|
- 📝 **Form Filling** - Fill out forms automatically
|
|
|
- 🖱️ **Point and Click** - Click buttons, links, and elements
|
|
|
- ⌨️ **Text Input** - Type text into fields
|
|
|
- 📜 **Page Scrolling** - Scroll through content
|
|
|
|
|
|
## How to Use
|
|
|
|
|
|
1. Enter a task in natural language (e.g., "Search for the latest news about AI")
|
|
|
2. Click "Run Task" and watch the agent work!
|
|
|
3. View the screenshots to see each step the agent takes
|
|
|
|
|
|
## Powered By
|
|
|
|
|
|
- **Microsoft Fara-7B** - Vision-Language Model for computer use
|
|
|
- **Playwright** - Browser automation framework
|
|
|
- **Modal** - Model hosting and inference
|
|
|
|
|
|
## Links
|
|
|
|
|
|
- [GitHub Repository](https://github.com/microsoft/fara)
|
|
|
|
|
|
## License
|
|
|
|
|
|
MIT License
|
|
|
|
