DigUp

DigUp is an open-source Mac app by Abdur Rahim (GitHub user ARahim3) that finds files by what's in them. Pick the folders to search and it reads the pictures, screenshots, PDFs, documents, recordings and videos inside; then press ⇧⌘Space in any app and describe what you want in your own words, such as "the zebra in a video" or "the clause about pets in the lease". It opens the match, jumping to the moment in a video or recording and to the page in a PDF. The AI part is Google DeepMind's multimodal embedding model EmbeddingGemma 2, running on your Mac; the README explains that it places text, images, audio and video in one shared space, so a single description can be compared against video frames and PDF pages alike. Exact strings such as invoice numbers and error codes are matched as written too. What sets it apart is searching across file types and languages: the README says an English search can find a note written in Bengali or Arabic. There is also an optional code search, off by default.

Who it's for:people with lots of photos, scans, meeting recordings or videos on their Mac who remember what's in a file but not what it's called.

Before you start: you need an Apple Silicon Mac on macOS 14 or later; the README says Windows and Linux versions are on the way. Install the DMG from GitHub Releases or use Homebrew. The first launch downloads an 865 MB model, and the first indexing pass takes from a few minutes to about an hour. The README says it reads only the folders you choose, never changes or uploads files, goes online only to download the model and check for updates once a day (which you can turn off), and needs no account; AI Inspiration World has not verified these privacy claims. The README also notes that languages written without spaces, such as Chinese and Japanese, have limits for exact-word matching, while searching by meaning is unaffected. The first time it jumps to a moment in a video, macOS asks whether DigUp may control QuickTime Player. It's free and MIT-licensed.

Based on the project's public documentation. Not tested by AI Inspiration World.