https://github.com/gaiborjosue/pdf-gpt

A GPT-like chat bot that will help you answer questions about multiple pdf files!
https://github.com/gaiborjosue/pdf-gpt

Last synced: 5 days ago
JSON representation

A GPT-like chat bot that will help you answer questions about multiple pdf files!

• Host: GitHub
• URL: https://github.com/gaiborjosue/pdf-gpt
• Owner: gaiborjosue
• License: mit
• Created: 2023-08-13T22:30:40.000Z (over 1 year ago)
• Default Branch: main
• Last Pushed: 2024-02-09T02:59:53.000Z (9 months ago)
• Last Synced: 2024-02-09T03:38:15.255Z (9 months ago)
• Language: Python
• Homepage: https://apdfgpt.streamlit.app
• Size: 7.91 MB
• Stars: 3
• Watchers: 3
• Forks: 1
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
# PDF GPT AI assistant

The PDF GPT website that allows users to upload multiple PDF files and ask questions about their contents using the GPT-3.5 conversational AI model. This extension leverages LangChain and OpenAI's GPT-3.5 to provide insightful answers to user queries based on the uploaded PDFs.

## Quick Demo:

![](https://github.com/gaiborjosue/PDF-GPT/blob/main/resources/streamlitdemo.gif)

## Features

- Upload multiple PDF files and extract their text content.

- Interact with the GPT-3.5 conversational AI model to ask questions about the uploaded PDFs.

- Seamless integration with Streamlit and LangChain libraries.

- User-friendly interface with responsive design.

- Gives 5 auto-generated Q/A so you can understand better the documents before-hand.

## How to Use

1. **Access the App**: To use the PDF GPT - MultiPDF Assistant, simply visit the web app at [https://apdfgpt.streamlit.app](https://apdfgpt.streamlit.app) using a web browser.

2. **PDF Upload**: In the app interface, navigate to the "Your Documents" section. Use the "Upload your PDFs here" button to upload one or more PDF files.

3. **Processing**: After uploading PDFs, click the "Process" button to extract text content, process it, and set up the conversational AI chain. This will also autogenerate 5 Q/A's.

4. **Asking Questions**: Use the "Ask your question here..." input field to ask questions related to the content of the uploaded PDFs.

5. **Responses**: The app will provide responses to your questions based on the uploaded PDFs' content and the AI model's analysis.

6. **About and Feedback**: The app also provides an "About" section with information about the PDF GPT - MultiPDF Assistant and a "Feedback" section where you can provide feedback or report issues.

## Dependencies

- Streamlit: Python library for creating web applications.

- LangChain: Library for processing and analyzing text data.

- OpenAI's GPT-3.5: AI model for natural language processing.

## Acknowledgments

- This project utilizes Streamlit, LangChain, and OpenAI's GPT-3.5. Special thanks to the authors and contributors of these libraries.

- I took inspiration from [Alejandro AO](https://www.youtube.com/@alejandro_ao)

## Feedback and Support

If you encounter any issues or have suggestions for improvement, please feel free to [open an issue on GitHub](https://github.com/gaiborjosue/PDF-GPT/issues).

## License

This project is licensed under the MIT License - see the [LICENSE](LICENSE) file for details.