https://github.com/ashad001/ultimaterag

An AI-based application leveraging Gemini/OpenAI and JinaAI embeddings with a FastAPI backend and Svelte frontend. The app can read PDFs, maintain webpage memory, and facilitate interactive chat with websites, webpages, and PDFs.
https://github.com/ashad001/ultimaterag

ai embeddings gemini jina llama-index llms machine-learning openai pdf-search rag vector-store web-search

Last synced: 5 days ago
JSON representation

An AI-based application leveraging Gemini/OpenAI and JinaAI embeddings with a FastAPI backend and Svelte frontend. The app can read PDFs, maintain webpage memory, and facilitate interactive chat with websites, webpages, and PDFs.

• Host: GitHub
• URL: https://github.com/ashad001/ultimaterag
• Owner: Ashad001
• Created: 2024-07-01T07:41:30.000Z (5 months ago)
• Default Branch: main
• Last Pushed: 2024-07-30T07:06:59.000Z (4 months ago)
• Last Synced: 2024-07-30T10:19:14.955Z (4 months ago)
• Topics: ai, embeddings, gemini, jina, llama-index, llms, machine-learning, openai, pdf-search, rag, vector-store, web-search
• Language: Python
• Homepage:
• Size: 1.81 MB
• Stars: 0
• Watchers: 3
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# UltimateRAG

This project is an AI-based application that leverages Gemini/OpenAI and JinaAI embeddings on a FastAPI backend. The application is capable of reading PDFs, maintaining webpage memory, and facilitating interactive chat with websites, webpages, and PDFs.

## Project Description

The AI-based application aims to provide an interactive and intelligent interface for users to engage with various digital content. The key features of the application include:

- **PDF Reading**: The application can read and understand the content of PDF documents.

- **Webpage Memory**: The application retains memory of interactions on webpages, allowing for a seamless user experience.

- **Interactive Chat**: Users can chat with the application about websites, webpages, and PDFs, receiving intelligent and context-aware responses.

## Project Structure

The project is divided into two main parts: the backend and the frontend.

### Backend

- **Framework**: FastAPI

- **AI Models**: Gemini/OpenAI, JinaAI embeddings

- **Capabilities**: Reading PDFs, webpage memory, interactive chat

### Frontend

- **Framework**: Svelte

- **UI/UX**: Responsive and user-friendly interface for interacting with the AI-based functionalities

## To-Do List

### Backend

1. **Set up FastAPI backend**

   - Install FastAPI and create the initial project structure

2. **Integrate AI Models**

   - Configure and integrate Gemini/OpenAI models

   - Set up JinaAI embeddings

3. **Implement PDF Reading Capability**

   - Develop endpoints to upload and read PDF documents

4. **Develop Webpage Memory Feature**

   - Create mechanisms to store and retrieve webpage interactions

5. **Enable Interactive Chat**

   - Implement chat functionality using AI models

### Frontend

1. **Set up Svelte Frontend**

   - Install Svelte and create the initial project structure

2. **Design User Interface**

   - Create a responsive and intuitive UI for interacting with the backend functionalities

3. **Implement PDF Interaction**

   - Develop frontend components to upload and display PDF content

4. **Develop Webpage Memory Interface**

   - Create frontend components to visualize and interact with stored webpage memories

5. **Enable Chat Interface**

   - Implement a chat interface to communicate with the backend

## Getting Started

### Prerequisites

- **Backend**: Python, FastAPI, Gemini/OpenAI, JinaAI

- **Frontend**: Node.js, Svelte