🦙 LLM Reader

A web-based chat interface for local Ollama models with PDF upload, image analysis, and website content retrieval capabilities.

✨ Features

• 💬 Real-time Chat - Stream responses from local Ollama models
• 📄 PDF Analysis - Upload PDFs and ask questions about their content
• 🖼️ Image Analysis - Upload images and analyze them with vision-capable models
• 🌐 Website Retrieval - Extract and analyze content from any website using Jina Reader API
• 🔄 Clean Interface - Simple conversation management and reset
• 🚀 Fast - Uses gemma3:4b model for quick responses with text, image, and web content support

🚀 Quick Start

Prerequisites

1. Install uv - Fast Python package manager
2. Install Ollama
3. Start Ollama: ollama serve
4. Pull models:
   • For text and image chat: ollama pull gemma3:4b

Installation & Run

cd LLM_Reader
# Run with default model gemma3:4b
uv run gradio_chat.py
# Run with specific model
uv run gradio_chat.py --model gemma3:1b

Open http://localhost:7860 in your browser (or your custom host:port).

🎯 Usage

• Chat: Type messages and get streaming responses
• Upload Files: Select "PDF Document", "Image", and/or "Website URL" to show upload options
• PDF Upload: Upload PDFs to analyze documents and ask questions about content
• Image Upload: Upload images for analysis (supported by default gemma3:4b model or specialized vision models)
• Website Retrieval: Enter any website URL to extract and analyze its content using Jina Reader API
  • Works without API key (with rate limits)
  • Optional Jina API key for better performance and higher rate limits
• Reset: Clear conversation and uploaded content
• Model Selection: Use --model parameter to specify which Ollama model to use
• Auto Cleanup: Model is automatically unloaded when stopping the interface (Ctrl+C)

🔧 Configuration

Command Line Options

uv run gradio_chat.py --help

Available options:

• --model MODEL_NAME - Specify Ollama model (default: gemma3:4b)
• --host HOST - Set host address (default: 127.0.0.1)
• --port PORT - Set port number (default: 7860)

📦 Dependencies

• gradio>=4.0.0 - Web interface framework
• ollama>=0.3.0 - Ollama Python client
• pymupdf4llm>=0.0.5 - PDF processing and markdown conversion
• requests>=2.25.0 - HTTP library for Jina Reader API calls

🌐 Jina Reader API

The website retrieval feature uses the Jina Reader API to extract clean, readable content from web pages.

📄 License

This project is open source and available under the MIT License.