Introduction
Manthana AI draws inspiration from the ancient Sanskrit story of Samudra Manthana (समुद्र मंथन) - the churning of the cosmic ocean. Just as the devas and asuras churned the cosmic ocean to extract amrita (the nectar of immortality), Manthana AI churns through your documents to extract valuable insights and knowledge.
What is Manthana AI?
Manthana AI is a powerful document analysis tool that combines the capabilities of local Large Language Models (LLMs) with advanced Retrieval-Augmented Generation (RAG) technology. It enables you to have meaningful conversations with your documents while keeping your data secure and private by running entirely on your local machine.
Key Features
Local First: All processing happens on your machine, ensuring complete privacy and data security
Multi-Document Analysis: Upload and analyze multiple documents simultaneously
Context-Aware Responses: Get precise answers that incorporate context from your documents
Smart Chunking: Advanced document segmentation for optimal knowledge retrieval
Intuitive Interface: Clean, modern UI inspired by the cosmic churning theme
Real-time Processing: See the churning process as your documents are analyzed
Flexible File Support: Support for PDF, DOCX, and TXT files
Architecture
Manthana AI is built on a modern technology stack that ensures high performance and reliability:
Frontend: Built with Tauri and React, providing a native desktop experience
Backend: Python-based processing engine with FastAPI
Embeddings: Local embedding generation using Ollama's nomic-embed-text model
Vector Store: Efficient document chunk storage and retrieval using Qdrant
LLM Integration: Seamless integration with Ollama for local language model inference
Core Components
Kalasha (Knowledge Repository)
The Kalasha component, named after the vessel that held the cosmic ocean, serves as your document repository. It manages:
Document storage and organization
File type validation and processing
Real-time upload status tracking
Samudra (Vector Store)
The Samudra component, representing the cosmic ocean itself, handles:
Document chunking and embedding generation
Vector similarity search
Context retrieval for question answering
Amrita (Knowledge Extraction)
The Amrita component, named after the nectar of immortality, delivers:
Context-aware responses
Natural language question answering
Document-grounded insights
License and Usage
Manthana AI is released under the MIT License. You are free to use it for personal and commercial purposes. See our license for more details.
Last updated