Mew 🐈 : Local RAG implementation and personal assistant ! (= ФェФ=) **

Introduction

Mew is a local implementation of a Retrieval-Augmented Generation (RAG) solution, designed as a personal assistant powered by efficient gguf models. Using Ollama technology, Mew delivers a streamlined and private assistant experience that runs entirely on your local machine, ensuring privacy and quick responses without relying on external servers.

How It Works

Mew operates with three main components:

1. Backend - Handles core functionalities and orchestrates data flow.
2. Frontend - Provides an intuitive interface for user interactions.
3. Ollama Server - Powers the assistant with language model processing.

These components interact seamlessly, allowing for real-time responses from Mew.

Setup

Prerequisites

Before setting up Mew, ensure you have the following:

• A processor capable of running gguf models (refer to relevant webpages for model compatibility).
• A PostgreSQL 16 database for your files uploads, with pgvector extension.
• Node.js installed.
• Python installed.
• Visual Studio Code (or any other IDE) for an enhanced development experience.
• Compatible with Linux and Windows (Mac compatibility unknown).

Setting up the Environment

1. Clone the Repository: Clone this repository to your local machine.
2. Install Dependencies: Use requirements.txt to install the necessary packages.

   pip install -r requirements.txt

How to Use Mew

1. Download a gguf Model: Obtain a gguf model compatible with Ollama.
2. Place Model in the models Folder: Move the downloaded model into the models directory within your repository.
3. Start the Server: Run the following command to start the server

  python server_setup.py

Once the setup is complete, Mew will be up and running. Enjoy your new personal assistant!