Skip to content

A generative AI app that responds to Phasmophobia-related queries.

Notifications You must be signed in to change notification settings

ak0411/phasmo-chat

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

33 Commits
 
 
 
 
 
 

Repository files navigation

phasmo-chat

phasmo-chat is a generative AI project leveraging Retrieval-Augmented Generation (RAG) to answer questions about the game Phasmophobia by Kinetic Games.

Features

  • Client Interface: A user-friendly interface, similar to ChatGPT, where users can chat and ask questions about Phasmophobia.
  • Question Answering API: API that answers questions about Phasmophobia using a Large Language Model (LLM) integrated with RAG for enhanced accuracy and context-aware responses.
  • Retrieval-Augmented Generation (RAG): Enhances the response accuracy by retrieving relevant information before generating answers, ensuring that responses are contextually relevant and accurate.
  • Response with Source URLs: Each response provides clickable links to the original sources, enhancing transparency and trustworthiness.

image.png image.png

How It Works

The front end is a ChatGPT-like interface where users can type in questions about Phasmophobia. When a question is submitted, the front end sends a request (POST) to the back end, which processes the query using RAG and returns a generated response to the front end.

Disclaimer

This is a personal project for learning purposes in generative AI and RAG. This project is not affiliated with Kinetic Games, the creators of Phasmophobia. The information provided by phasmo-chat is sourced from the Phasmophobia Wiki.

Front-end Technologies

  • React: JavaScript library for building user interfaces.
  • TypeScript: Typed superset of JavaScript that compiles to plain JavaScript.
  • Vite: Build tool providing a fast and lean development experience.
  • Axios: Promise-based HTTP client for the browser and Node.js.
  • React Query: Manages fetching, caching, and updating asynchronous data in React.
  • Material UI: React component library for building user interfaces with pre-built components and styling solutions.

Back-end Technologies

  • LangChain: Framework for developing applications powered by language models.
  • OpenAI: Provides the language model for generating responses.
  • FastAPI: High-performance web framework for building APIs.
  • ChromaDB: Database for managing and querying embeddings.
  • Cohere: Assists with language understanding and generation.