New

Imagory 0.2 is out, now with a Linux build. Changelog

Download

Your photo library, offline and organised.

Browse your local photo folders and Google Takeout exports in one fast grid. Nothing is uploaded, and your photos are only ever read.

Download for Windowsv0.2.0Download for Linuxv0.2.0
All releases

Windows 10 & 11 · Linux x86_64 · Free and open source · No account

The Imagory main window: a sidebar with Library, People, Albums and Sources, and a grid of photo thumbnails grouped under month headers.The Imagory main window: a sidebar with Library, People, Albums and Sources, and a grid of photo thumbnails grouped under month headers.

Features

Imagory's first-run screen: add a folder of photos or an extracted Google Takeout export.
The first-run screen. Imagory only reads your photos, and nothing leaves your computer.

All your folders, one library

Add photo folders and extracted Takeout exports as sources. Imagory scans them in the background and shows everything in one grid, newest first.

  1. The photo grid, grouped under month headers, with the sources sidebar.The photo grid, grouped under month headers, with the sources sidebar.
    Grid with month headers
  2. The photo viewer with the info panel open, showing date, place, people and file.
    Viewer and info panel
  3. The Sources dialog listing a photo folder and a Google Takeout export with their photo counts.
    Folders and Takeout sources
  4. The Clean up Takeout wizard showing what it will copy and how much space it needs.
    Clean up Takeout

Fast with 100,000 photos

The first scan of a large library takes a while, so it runs in the background. After that, opening the library and rescanning it are quick.

  • Scans run in the background with progress, and photos appear as they are processed.
  • Cancel any time. The next rescan carries on where it stopped.
  • The window stays responsive: the heavy work runs in separate worker processes.
  • A file you moved inside a source keeps its place instead of showing up as missing and new.
TaskTargetMeasured
Photos in one scrolling grid100,000100,000
Open a scanned library to the first thumbnailsunder 1 s0.22 s
Rescan 100,000 unchanged photosunder 1 min6.3 s (95,000 photos)
Measured on Linux with the project's benchmark script and a generated library. Windows figures will follow.

Turn a messy Takeout into a tidy folder

Takeout puts a JSON file next to every photo, copies photos into each album folder, and sets file dates to when you unzipped it. Clean up Takeout makes a sorted copy, one file per photo, in YYYY/MM folders. How it works

Before Extracted Takeout export
  • Takeout
    • Google Photos
      • Photos from 2019
        • IMG_20190131_235959.jpg
        • IMG_20190131_235959.jpg.supplemental-metadata.jsonJSON
        • PXL_20190704_182311.jpg
        • PXL_20190704_182311.jpg.jsonJSON
      • Lisbon trip
        • IMG_20190131_235959.jpgduplicate
        • IMG_20190131_235959.jpg.jsonJSON
      • Trashnot copied
After Your output folder
  • Photos (cleaned)
    • 2019
      • 01
        • IMG_20190131_235959.jpg
      • 07
        • PXL_20190704_182311.jpg
    • .imagorymetadata
      • imagory.db
      • metadata.json
      • reports

  • Copies only

    It writes into an output folder you pick.

  • Every copy verified

    Checked against the original before it counts.

  • Metadata kept

    Dates, places and people go into metadata.json.

  • Original untouched

    The export is never changed. Deleting it is up to you.

Everything in one place

  • Exact duplicates shown once

    Identical files appear once in the grid, wherever they are in your sources. Moved files are recognised.

  • The right dates

    From Takeout, from EXIF (with its time zone when known) or from the filename, such as IMG_20190131_235959.

  • Search and filters

    Narrow the grid down by what you're looking for, including by person.

  • Albums

    Group photos into albums. Albums from a cleaned Takeout come along too.

  • People from Takeout

    The names tagged in Google Photos, read from the export. Imagory does no face recognition.

  • Viewer with zoom

    Open any photo, zoom in, and see its date, place, people and file in the info panel.

Free and open source

Version 0.2.0. No account, no sign-up, nothing to pay.

Windows

Imagory.exe

Runs on
Windows 10 and 11, 64-bit
Size
about 50 MB

A single file. Double-click to run, nothing to install.

Download for Windowsv0.2.0

Linux

Imagory-0.2.0-linux-x86_64.tar.gz

Runs on
x86_64, Ubuntu 22.04 or newer, X11 or XWayland
Size
about 50 MB

Unpack it, then run ./Imagory. Best effort: less tested than Windows.

Download for Linuxv0.2.0

Windows may warn about an unsigned app. How to run it

All releasesBuild from source

Questions

Something else? More in the FAQ.

Is it really offline?

Yes. The app makes no network connections: no telemetry, no accounts, no update checks. Your photos, the database, the thumbnails and the logs all stay on your computer.

Does Imagory change my photos?

No. It only reads them. It never changes, moves, renames or deletes a photo, and never writes anything into your photo folders. Clean up Takeout writes copies only into an output folder you pick.

Does it work with Takeout zips?

Not yet. Extract the zip files first, then add the extracted folder as a source. Zip support is planned.

Why does Windows warn me about it?

Imagory releases are not code-signed, so Windows SmartScreen may say “Windows protected your PC”. Click More info, then Run anyway. Some antivirus tools also flag unsigned single-file programs. You can build it yourself from source instead.

Where is my library stored?

In %LOCALAPPDATA%\Imagory on Windows and ~/.local/share/imagory on Linux. It holds the database, a thumbnail cache you can delete at any time, and the logs.

Does it find faces?

No. Imagory does no face recognition. The People section shows the names tagged in Google Photos, read from your Takeout export.

Is it free?

Yes. Imagory is free and open source under the MIT licence. There is no account and nothing to buy.