MemoryBox: A Privacy-First Screenshot Search Tool Powered by Local AI
MemoryBox is a privacy-focused screenshot search tool that lets students, developers, and researchers find images using natural-language queries. The current browser demo requires manual uploads and relies on local OCR, the Gemma 3 4B vision model, and local embeddings for search. Its planned desktop version would index selected screenshots automatically, with user permission, and run open-weight models on the user's device. The developer plans to sell it as a one-time desktop license. The project was submitted to the Hacktoberfest Open-Source AI Challenge.
This is an AI-generated summary. ShortSingh links to the original source for the complete article.

Discussion (0)
Log in to join the discussion and vote.
Log in