Recall.OS
AI search for all your files. Nothing leaves your computer.
What it does
Most AI tools make you upload files to their servers and hope they answer correctly. Recall.OS flips everything: runs locally, auto-watches folders, and doesn't hallucinate because it's searching YOUR files not generating answers. Every response cites exact sources—page 12, timestamp 3:42, not "trust me bro." 25 documents free forever. Unlimited? $30 once, not $20/month forever. They own your data. You own the tool. First AI that treats your files like they're actually yours.
Does the same job
all alternatives →
- RRRem: Remember Everything (open source)2023 · github.com · ▲555
An open source approach to locally record everything you view on your Apple Silicon computer. Note: Relies on Apple Silicon, and configured to only produce Apple Silicon builds. I think the idea of recording everything you see has the potential to change how we interact with our computers, and believe it should be open source. Also, from a privacy / security perspective, this is like... pretty scary stuff, and I want the code open so we know for certain that nothing is leaving your laptop. Even logging to Sentry has the potential to leak private info.
Find images by what they show, AI-rename them, and sort them easily20d ago · apps.microsoft.com · ▲10ImageSage finds the image you remember.

It is a private, local AI gallery. Point it at the folders you choose and it reads them in place, building an index that stays on your PC. That gives you two things Windows file search cannot: search that understands what is inside your images, and AI
More ai this month
the category →
I trained a 125M-parameter transformer to autocomplete piano performances in real time (~108 notes/sec on an iPhone 15). The idea is basically GitHub Copilot or Tabnine, except instead of prompting it with code, you prompt it by playing a few notes on a MIDI piano. The model then continues what you played, entirely on-device. The app is free if anyone wants to try it. Happy to answer questions about the model, training, Core ML, or the many things that didn't work.
AI · 17d ago · simedw.com
Astute▲585Automate your B2B brand going viral, with new media creators
AI · 18d ago · company-app.joinastute.com
Hey HN, Henry from Cactus here! We previously released Cactus Needle, a 14MB agentic LLM for tool call, device use, and structured extraction for phones, wearables, smart homes, small robots and microcontrollers. We got really great feedback here, and have now incorporated the suggestions to release Needle 2. The whole model is a single 14MB binary that runs a full session in 28MB of RAM; 45m parameters at 2bit compression. Needle hits 500 tokens/sec decode speed on a Raspberry Pi 5, sits between 400-1,500 tokens/sec on VR devices like Meta Quest 3S and Apple Vision Pro, and ranges…
AI · 27d ago · cactuscompute.com
