My friend Sami and I recently built Vizly, a Mac application that allows anyone to query their databases using plain English.
Vizly is built on Llama 2, llama.cpp, and runs fully on-premise.
We are running two Llama models, one for natural language to SQL translation, and another that uses the results from the SQL to render visualizations. That means there are no external APIs and all the AI models are running locally on your MacBook.
We tried to make Vizly very easy to share as well. Every Vizly instance creates a share link that can be accessed by anyone on the same network as you. Just send the share link to anyone on the same network and they will be immediately able to run AI-powered queries, hosted from your device.
Would love if you could try it out and give us any feedback!
My friend Krish and I recently built MiniMe, a GPT-4 text assistant that learns from your messages and can be used to reply on your behalf on any platform (iMessage, WhatsApp, Twitter, LinkedIn, etc).
You can toggle AI autoreply on for specific conversations and be notified if there's a specific task you need to action.
Will the future be AI chatbots talking to other AI chatbots?