ComingUp ComingUp

About

Hey everyone. I wanted to share a project I’ve been working on: Privami AI. It’s a fully open-source mobile application that lets you run LLMs locally on your phone hardware and has web search/RAG using the DuckDuckGo API. Why I built this? I wanted a simple AI assistant for everyday use that everything I ask stays inside my phone and not go trough any external APIs or similar. But I also needed to have access to up-to-date information, so I added web search using DuckDuckGo API, which respects...

Comments (4)

Amir Jakubowski Amir Jakubowski 3 months ago

Which models actually run without lagging on midrange phones?

Laney Crist Laney Crist 3 months ago

how bad does battery drain running 7b locally

Juliet Hoppe Juliet Hoppe 3 months ago

whats the inference engine running underneath?

Jaime Schoen Jaime Schoen 3 months ago

local inference on phone hardware is a fun thermal problem