GenericAI: A ReAct-Based Intelligent Assistant

GenericAI is a versatile AI assistant developed using the ReAct (Reasoning and Acting) framework. It leverages LangChain for coordinating language model interactions and Gradio for an intuitive interface. With a variety of integrated tools, GenericAI is capable of reasoning through complex queries and executing actions to deliver accurate and well-rounded responses.

Table of Contents

• Features
• Demo
• Installation
  • Prerequisites
  • Setup
• Configuration
• Usage
  • Running the Application
  • Interacting with GenericAI
• Tools
• Contributing
• License
• Acknowledgements

Features

• ReAct Framework: Uses reasoning and acting to address complex queries.
• Integrated Tools: Includes Python REPL, Terminal, Weather API, Web Browsing, PDF Summarization, Image Description, and more.
• Chat History Management: Tracks conversations with MongoDB for contextual responses.
• User-Friendly Interface: Built with Gradio for easy interaction.
• Extensible Architecture: Add or modify tools with minimal effort.

Demo

Run the application locally to experience GenericAI in action. Submit questions, provide files, and see how it intelligently uses tools to respond.

Installation

Prerequisites

• Python 3.12: Install from python.org.
• MongoDB: Use MongoDB Atlas for cloud or install locally.
• Git: Required to clone the repository.

Setup

1. Clone the Repository

   git clone https://github.com/shaswata56/GenericAI.git
   cd GenericAI

2. Create a Virtual Environment

   Use a virtual environment to manage dependencies.

   python3 -m venv venv
   source venv/bin/activate  # On Windows: venv\Scripts\activate

3. Install Dependencies

   pip install --upgrade pip
   pip install -r requirements.txt

Configuration

1. Environment Variables

   This repository includes a pre-configured .env file. However, when you use it, ensure that you add .env to your .gitignore to avoid exposing sensitive information. Below is the default content of the .env file:

   OPENROUTER_API_KEY="your-api-key"
   TOGETHER_API_KEY="your-api-key"
   OPENWEATHERMAP_API_KEY="your-api-key"
   TAVILY_API_KEY="your-api-key"
   MONGODB_URL="mongodb://localhost:27017/"
   LLM_PROVIDER="ollama"
   LLM_MODEL="llava:13b"
   VLM_PROVIDER="ollama"
   VLM_MODEL="llava:13b"

   Update the values in the .env file with your actual API keys and configuration as needed.

2. API Keys

   Obtain API keys for tools like OpenWeatherMap, Tavily, and PubMed, and update the .env file accordingly.

Usage

Running the Application

Activate the virtual environment if it isn't active:

source venv/bin/activate  # On Windows: venv\Scripts\activate

Start the application:

python GenericAI.py

The Gradio interface will provide a local URL (e.g., http://127.0.0.1:8080). Open the URL in your browser to interact with GenericAI.

Interacting with GenericAI

1. Ask a Question: Enter your query in the text box.
2. Upload a File (Optional): Upload a file (e.g., PDF, image) for processing.
3. View Response: The response will appear in the markdown output area.

Example Queries:

• "Summarize the attached PDF."
• "What is the weather like in New York?"
• "Describe the image in the uploaded file."
• "Count the number of vowels in the word 'OpenAI'."

Tools

GenericAI includes various tools for enhanced functionality:

Core Tools

• Python REPL: Execute Python commands.
• Terminal: Run shell commands.
• Weather: Fetch weather data using OpenWeatherMap.
• Search: Perform web searches.
• Calculator: Solve math problems.
• Wikipedia: Access general knowledge.
• YouTube: Retrieve YouTube video links.
• Arxiv: Search for scientific articles on arXiv.
• DateTimeNow: Display the current date and time.

Specialized Tools

• LetterInWordCounter: Count occurrences of a letter in a word.
• WebpageBrowser: Summarize webpage content.
• PubMed: Query medical information.
• RecallMemory: Retrieve past conversations.
• VowelCounter: Count vowels in a string.
• TotalLetterCounter: Count letters in a string.
• PdfSummarizer: Summarize PDF files.
• ImgDesc: Describe images.
• ImgDescOnline: Describe online images.
• TxtFileReader: Read text files.
• TxtFileWriter: Overwrite text files.
• TxtFileAppender: Append to text files.
• PdfReader: Read PDF files.
• PdfReaderOnline: Read online PDFs.
• Notepad: Simple notepad functionality.

Adding Custom Tools

Define new tools and register them in the get_tools function in GenericAI.py. Ensure each tool adheres to input-output standards for smooth integration.

Contributing

Contributions are welcome! Follow these steps to contribute:

1. Fork the Repository

   Click the "Fork" button at the top right of this page.

2. Clone the Forked Repository

   git clone https://github.com/shaswata56/GenericAI.git
   cd GenericAI

3. Create a Feature Branch

   git checkout -b feature/YourFeatureName

4. Commit Your Changes

   git commit -m "Add Your Feature"

5. Push to the Branch

   git push origin feature/YourFeatureName

6. Open a Pull Request

   Navigate to your forked repository on GitHub and click "Compare & pull request".

License

This project is licensed under the MIT License.

Acknowledgements

• ReAct: Synergizing Reasoning and Acting in Language Models for enabling step-by-step reasoning and acting to solve complex problems.
• LangChain for its powerful framework for language model applications.
• Gradio for its user-friendly interface creation tools.
• MongoDB for scalable storage of chat histories.

---