This repository contains various applications built based on Large Language Models (GPT-4, ChatGPT, and/or LLAMA).
- Meeting Summarizer
- PDF Summarizer
- POC Builder
- Wikipedia LLM-Interface
Summarizes a meeting or a conversation based on an input transcript.
Upload a PDF document and summarize it in language of choice.
Generates code for any given use-case using GPT-4. Enables users to specify their requirements in natural language, and creates a webapp using Streamlit. Provided an experimental solution to quickly build POCs using LLMs.
Examples of use-cases:
# Create an app using which users can upload pdf documents, that are then summarized by chatgpt
# Create a sample streamlit app with dataframe, static table, line chart, and scattarplot
# Create an app using which a user can upload pdf documents. There should be a button to get a summary. There should a text area where a user can add some natural language instructions. These instructions should be processed by AI and then used to create bar or pie charts, if the user clicks on appropriate buttons.
A conversational knowledge bank built by combining the language capabilities of GPT-4 with the You.com search engine and Wikipedia API. This Streamlit-based webapp provides a natural language interface to Wikipedia. Enables users to ask questions via voice or have a multi-turn interaction via a text-based chat.
Below is an illustration of various components:
Streamlit app interface:
