Your personal AI-powered book assistant that runs completely offline
Text2LLM is a Windows desktop application that transforms how you interact with PDF documents. Upload any book or paper, ask questions in plain English, and get instant answers with exact page citations—all powered by AI that runs locally on your machine.
- 100% Local - All processing happens on your machine
- No Cloud Dependencies - Your documents never leave your computer
- Completely Offline - Works without internet after initial setup
- Plain English Queries - Ask questions like you're talking to a person
- Context-Aware Answers - Get relevant responses based on your document
- Page Citations - Every answer includes exact page references
- Large Document Support - Handle 900+ page books effortlessly
- Fast Responses - Get answers in 5-10 seconds
- Setup Wizard - Guided installation for first-time users
- Clean Interface - Intuitive design focused on productivity
- No Configuration - Works out of the box
Size: ~25 MB | Platform: Windows 10/11 | License: MIT
Check regularly for new releases!
-
Download Text2LLM (25 MB)
-
Install Ollama (600 MB)
- Setup wizard will guide you
- Download from: https://ollama.com
- Simple 1-click installer
-
Download Models (2.2 GB - automatic)
- Text2LLM downloads these for you
- llama3.2:3b + nomic-embed-text
- Takes 5-10 minutes
Total Size: ~2.5 GB
Total Time: ~10-15 minutes
Then: Use forever, completely offline!
- Works offline
- Cache books once, query forever
- Launch Text2LLM (after setup)
- Click "Load PDF Book"
- Select your PDF file
- Wait for indexing (first time only: 10-15 mins for large books)
- Ask questions in the query box
- Get answers with page citations!
"What is the main argument of chapter 3?"
"Summarize the methodology section"
"What does the author say about climate change?"
"List all the conclusions mentioned in the book"
"What are the key findings on page 45-50?"
- Python 3.14 - Application framework
- Tkinter - Cross-platform GUI
- Ollama - Local LLM inference engine
- llama3.2:3b - Advanced Reasoning Language model (3B parameters)
- nomic-embed-text - Text embedding model
- RAG Architecture - Retrieval-Augmented Generation
- PyPDF2 - PDF parsing and text extraction
- NumPy - Vector operations and similarity search
- Requests - HTTP client (setup only, checks Ollama server status)
- PyInstaller - Executable packaging