Spaces:
Sleeping
Sleeping
| title: "🤖 Agentic Browser" | |
| emoji: "🤖" | |
| colorFrom: "blue" | |
| colorTo: "purple" | |
| sdk: streamlit | |
| app_file: app.py | |
| pinned: false | |
| license: "mit" | |
| tags: | |
| - ai | |
| - browser | |
| - agent | |
| - llm | |
| - streamlit | |
| short_description: An autonomous browser agent for web-based tasks | |
| # 🤖 Agentic Browser | |
| A powerful AI-powered browser agent that can help you with web-based tasks using local language models. This application provides a chat interface to interact with various open-source language models. | |
| ## 🌟 Features | |
| - **Local AI Models**: Run models directly in your browser | |
| - **Multiple Models**: Choose between different models based on your needs | |
| - **Chat Interface**: Easy-to-use conversational interface | |
| - **Real-time Responses**: Streaming text generation with visual feedback | |
| - **Customizable**: Adjust parameters like temperature for different response styles | |
| ## 🚀 Quick Start | |
| 1. Select a model from the sidebar | |
| 2. Click "Load Model" to initialize the selected model | |
| 3. Start chatting in the main interface | |
| 4. Adjust temperature for different response styles | |
| ## 🤖 Available Models | |
| - **TinyLlama**: Fast and lightweight, great for quick responses | |
| - **Mistral-7B**: More powerful conversational model | |
| ## ⚙️ Settings | |
| - **Model Selection**: Choose from available models | |
| - **Temperature**: Control response creativity (0.1 = deterministic, 1.0 = creative) | |
| - **Real-time Loading**: Models are loaded on-demand for efficiency | |
| ## 🛠️ Technical Details | |
| This application uses: | |
| - Streamlit for the web interface | |
| - Hugging Face Transformers for model loading | |
| - PyTorch for model inference | |
| - Optimized for both CPU and GPU inference | |
| ## 📄 License | |
| This project is licensed under the MIT License. | |