Skip to content

scmpevatrons/ai-chatverse

Repository files navigation

Introducing AI Chatverse: A Playground for Chat Models

Dive into the world of conversational AI with AI Chatverse, a free and open-source application tailored for enthusiasts, developers, and AI aficionados alike. Built upon the robust foundation of Streamlit, AI Chatverse offers a user-friendly chat interface that seamlessly integrates with any chat model, be it from Langchain or a custom-built solution of your choice.

What sets AI Chatverse apart is its feature:

  1. To experience multi-agent collaboration interactions with MetaGPT and more (in the future) to accomplish a task.
  2. the ability to dynamically craft a 'session model'. These session models allow users to modify and experiment with existing models in real-time. Whether you're keen on tweaking the system prompt, adjusting hyperparameters like temperature, or just wish to test the waters with chatbot models, AI Chatverse offers a playground for all your experimentation needs.

Multi Agent Collaboration Collaboration Settings Chat with Multi Agents Chat with the model View your models Edit or Create Derivative Models

Table of Contents

  1. Features
  2. The Architecture
  3. Quickstart
    1. Visit the hosted app
    2. Use the official docker image
    3. Build the docker image from source
  4. Documentation
  5. Maintainers
  6. Attribution

Features

  • Open Source and Free The project has an opensource license.

  • Easy Integration with langchain its very easy to integrate a model available in langchain.

  • Doesn't store data We do not store any of your conversations, your messages are stored in streamlit sessions i.e once you close the App the data is gone!.

    Note: The model you use might store the data on it's end. Look into the specific model for more details.

  • Self-Hosted: The application can be run in any environment which can run python application, we have dockerised the container so you can use that image as well.

The architecture

AI Chatverse Architecture

Quickstart

The best way to get started with the application is by visiting the application hosted in streamlit community cloud here.

Note: LLAMA2 uses llama cpp and needs a aritifact, it can only be run locally.

PS: Please configure the installation according to this link

Running from docker locally via official docker image

The application can be run locally using docker by running the following command

docker run --rm -it -p 8501:8501 pevatrons/ai-chatverse

Running the latest version of the app from source in docker

  1. Clone the repository.
  2. Build the image by running
    docker build -t ai_chaverse .
    
  3. Run the application by
    docker run --rm -it -p 8501:8501 ai_chaverse:latest
    
  4. Visit localhost:8501

Documentation

The documentation for the project is available in readthedocs.

Maintainers

The application is created by Pevatrons.

We welcome collaboration and your contribution for this project.

Attribution

Some of the logos in the view models have been generated using Imagine Art be mindfull of this if you are using it for commerical purposes.