Lap is an open-source desktop photo manager for large local photo libraries, built to browse, search and clean up a collection that never leaves your own disk. It is aimed at people with years of family photos sitting on a laptop or an external drive — anyone who wants a searchable, organized library without first uploading it to a cloud service. It runs on macOS, Windows and Linux, is licensed under GPL-3.0, and is free to use.
What it does
The README describes Lap as a privacy-focused alternative to cloud photo services, and the feature set follows from that: no forced upload, local AI search, a folder-first workflow.
The main things it offers:
- Search that runs locally. You can look for old photos by what they contain, and the model doing the matching runs on your machine rather than in someone else's data center.
- Face grouping. Photos of the same person are clustered together so you can page through a person instead of a date range.
- Duplicate cleanup. It detects duplicate images and helps clear them out, which is usually the first thing a decade-old library needs.
- Large libraries. The project targets collections past 100,000 photos and aims to keep browsing smooth at that size.
- Offline operation. Nothing about the core workflow depends on a network connection.
The framing throughout is a family album you actually keep, rather than a professional editing or retouching tool. Lap organizes and finds photos; it is not an image editor.
How it works
Lap is a desktop application you install, not a server you deploy. There is no container to run, no reverse proxy to configure and no account to create — you download an installer for your platform and point the app at your photos.
Because it is folder-first, the directory structure already on your disk stays meaningful. The app works with the folders you have rather than sucking everything into an opaque catalog and renaming files underneath you, which matters if you also touch those photos with other software or back them up with an ordinary file-sync tool.
Indexing, search, face clustering and duplicate detection all happen on the local machine. That is the whole point of the design, and it is also the honest trade-off: the first pass over a very large library is work your own CPU has to do, rather than work a cloud provider does for you while you wait.
The interface is written in Vue and packaged into native installers for each supported platform, so the app looks and installs like a normal desktop program on all three operating systems.
Getting started
Open the project's latest release page and download the file that matches your system. The README lays the options out directly:
- macOS, Apple Silicon or Intel — a
.dmgfor each architecture, notarized by Apple, so it opens without a Gatekeeper fight. - Windows 10 or 11, x64 or ARM64 — an
.msiinstaller. These are unsigned, so SmartScreen may block the download; the README tells you to click Keep anyway. - Linux — a package is listed on the same release page.
The README is also translated into several languages, which is a reasonable signal that the project expects non-technical users from many places to install it directly rather than build from source.
When to use it / when not
Reach for Lap when your library lives on your own disk, you want it searchable, and you would rather not upload family photos anywhere. It is a good fit if your collection is big enough that manual folder browsing has stopped working, if duplicates have piled up across years of imports, or if you want faces grouped without a cloud account attached to them.
It is the wrong tool if what you actually want is sync. Nothing in the materials describes a mobile app, cross-device syncing, shared albums or a web interface you can reach from a phone — this is a program on one computer, managing files on that computer. If your real problem is getting photos off a phone and onto three devices, Lap does not solve it. GPL-3.0 is also worth noting if you had any plan to build something closed on top of it, and Windows users should expect the unsigned-installer warning.
Alternatives
The comparison the project makes for itself is against cloud photo services, where search and face grouping are free but conditional on uploading everything. The other shape of answer is a self-hosted photo server: you run software on a machine you control, it serves a web interface, and your devices sync to it. That gives you multi-device access at the cost of running and maintaining a server. Lap deliberately takes neither path — it is a plain desktop app, which is the lowest-effort option of the three and the only one that keeps working with no network at all.
Take Lap seriously if you are the person in the family who ended up holding everyone's photos. It has drawn about 2,800 stars since its first commit in August 2024 and is still being pushed to, which is a fair sign of momentum for a young desktop project. The pitch is narrow and it is honest about it: search, faces and duplicates for a big local library, on your machine, for free.