Inspiration
We were working with a research group that was training AI on different GPT plugins and APIs, so we were inspired to build our own tool with an interface we created.
What it does
It integrates and uses different GPT plugins and APIs seamlessly in a singular prompt without having to install and switch between plugins.
How we built it
We built it using a LLAMA trained on 1000 A100s tailored to 6 specific APIS. We built an API in Python and an interface in react to communicate with the model.
Challenges we ran into
It was challenging to develop the API to communicate between the react frontend and stream responses from the AI.
Accomplishments that we're proud of
We are proud of integrating the different components of our application.
What we learned
We learned how to use LangChain. This was also our first time building a project in react.
What's next for AutoPlugin
We wanted to integrate more APIs and plugins.
Built With
- chatgpt
- deepspeed
- langchain
- llama
- python
- react.js
Log in or sign up for Devpost to join the conversation.