19.2k stars · GPL-2.0 · v1.32.12 (2026-09-12), read from PyPI; the GitHub sidebar still showed v1.32.9 on the morning of the edition · Track this in Scout
A command-line downloader for image galleries and collections across hundreds of sites, with configurable filename templates and three ways to authenticate.
▶Repo detailsthe review · specs · pros & cons · install
What it is
A command-line program that downloads image galleries and collections from several hundred sites. It handles paging, file naming, skipping what you already have, and keeping the metadata, and you configure all of that in one JSON file.What it is good for. Anyone researching what competitors publish. Edition 6 gave you yt-dlp (#4) for video and left the pictures uncovered, and pictures are where a lot of the signal is. For a YouTube channel it collects every thumbnail a competitor has ever used, which is the cheapest thumbnail research available. Download for research and keep it to research: somebody else's images are still theirs.
- Several hundred sites are supported, and new ones arrive in almost every release.
- It records where each file came from, so your research stays traceable months later.
- It hands video addresses over to yt-dlp, so the two together cover pictures and video with one habit.
- GPL-2.0, so you cannot build a closed product around it.
- Many sites need your own login details written into the settings file, which leaves a secret sitting on disk. Lock that file down, or encrypt it with age (#4).
- Active development has moved to Codeberg. The GitHub page is now a mirror. Releases still appear here, and the version on PyPI was newer than the one GitHub was showing on the morning this was written, but issues and new code live at codeberg.org/mikf/gallery-dl.
yt-dlp/yt-dlpThe same command-line approach for video and audio instead of images, and gallery-dl hands video links to it rather than competing.
Track this in Scout- Bionus/imgbrd-grabber
The same bulk image downloading with a window you click instead of a command, aimed at imageboards, so it is friendlier and covers far fewer sites.
Track this in Scout - instaloader/instaloader
The same job for one site only, Instagram, and it goes deeper there with captions, stories and comments than a general tool does.
Track this in Scout
# Python 3.9 or newer python3 -m venv venv source venv/bin/activate pip install -U gallery-dl gallery-dl --version # Download a whole gallery into a tidy folder: gallery-dl -d ~/research "https://www.reddit.com/r/somesubreddit/" # See what it would do, without downloading anything: gallery-dl --simulate "https://www.reddit.com/r/somesubreddit/"
