Social Media Downloader (gallery-dl)

Download images and media from 100+ websites using gallery-dl. Supports Instagram, Reddit, Twitter/X, DeviantArt, Pixiv, 500px, Flickr, Pinterest, Bluesky, and more. Features: resolution filtering, duplicate detection, automatic file organization, and persistent config paths.

By EricRollei·Created 9 months ago·Updated 6 months ago· 5

output_dir
summary
download_count
success

◄url_list# Enter URLs here, one per line # Supports 100+ sites: Instagram, Reddit, Twitter/X, DeviantArt, Pixiv, 500px, Flickr, Pinterest, Bluesky, and more # Examples: # https://www.instagram.com/username/ # https://www.reddit.com/r/subreddit/ # https://twitter.com/username►

◄output_dir./gallery-dl-output►

◄url_file►

◄config_path►

◄cookie_file►

◄use_browser_cookiesfalse►

◄browser_namefirefox►

◄use_download_archivetrue►

◄archive_file./gallery-dl-archive.sqlite3►

◄skip_videosfalse►

◄filter_by_resolutiontrue►

◄min_image_width768►

◄min_image_height768►

◄extract_metadatatrue►

◄organize_filestrue►

◄instagram_includeposts►

◄download_timeout1800►

◄extra_options►

◄seed0►

CategoryDownloaders

Inputs (19)

Name	Type	Default	Description
url_list	STRING	# Enter URLs here, one per line # Supports 100+ sites: Instagram, Reddit, Twitter/X, DeviantArt, Pixiv, 500px, Flickr, Pinterest, Bluesky, and more # Examples: # https://www.instagram.com/username/ # https://www.reddit.com/r/subreddit/ # https://twitter.com/username	Enter URLs to download from, one per line. Supports Instagram, Reddit, Twitter/X, DeviantArt, Pixiv, 500px, Flickr, Pinterest, Bluesky, and 90+ other sites. Lines starting with # are ignored.
output_dir	STRING	./gallery-dl-output	Directory where downloaded files will be saved. Relative paths are relative to ComfyUI's output folder. Files are organized into subfolders by site and user.
url_fileopt	STRING		Optional: Path to a text file containing URLs (one per line). Useful for batch downloading from a prepared list. Leave empty to use url_list above.
config_pathopt	STRING		Path to gallery-dl.conf config file. This single file contains credentials for ALL sites (Instagram, Reddit, 500px, etc.) - gallery-dl auto-selects the right ones based on URL. Path is auto-saved for next time. Recommended: ./configs/gallery-dl.conf
cookie_fileopt	STRING		Path to exported cookie file (Netscape/JSON format). Only used when 'Use Browser Cookies' is OFF and you need site-specific cookies not in config. Usually leave empty - config file credentials work better. Path is auto-saved.
use_browser_cookiesopt	BOOLEAN	false	OFF (recommended): Use credentials from config file - most reliable. ON: Extract cookies directly from browser (Firefox works best; Chrome/Edge require admin rights and browser closed).
browser_nameopt	COMBO	firefox	Browser to extract cookies from when 'Use Browser Cookies' is ON. Firefox recommended - works without admin rights. Chrome/Edge need admin + browser closed.
use_download_archiveopt	BOOLEAN	true	ON (recommended): Track downloaded files in SQLite database to skip duplicates on future runs. OFF: Re-download everything, may create duplicates.
archive_fileopt	STRING	./gallery-dl-archive.sqlite3	Path to download archive database (SQLite). Tracks what's been downloaded to avoid duplicates. Delete this file to force re-downloading everything.
skip_videosopt	BOOLEAN	false	ON: Download only images, skip all video files. OFF: Download both images and videos. Useful when you only want still images from a mixed gallery.
filter_by_resolutionopt	BOOLEAN	true	ON: Skip images smaller than min_image_width/height (removes thumbnails, icons, low-res images). Videos are never filtered. Sites without dimension metadata (Bunkr, Cyberdrop) will download all images. OFF: Download all images regardless of size.
min_image_widthopt	INT	7680–8192	Minimum image width in pixels when filter_by_resolution is ON. Images narrower than this are skipped. Set to 0 to disable width filtering. Note: Sites that don't provide dimension metadata will download all images.
min_image_heightopt	INT	7680–8192	Minimum image height in pixels when filter_by_resolution is ON. Images shorter than this are skipped. Set to 0 to disable height filtering. Note: Sites that don't provide dimension metadata will download all images.
extract_metadataopt	BOOLEAN	true	ON: Save download metadata to JSON file (URLs, timestamps, file counts, errors). Useful for tracking what was downloaded. OFF: No metadata file created.
organize_filesopt	BOOLEAN	true	ON: Sort downloaded files into subfolders by type (images/, videos/, audio/, other/). OFF: Put all files in the main output folder without sorting.
instagram_includeopt	COMBO	posts	Instagram only: What content to download. 'posts' = feed posts, 'stories' = current stories (24hr), 'highlights' = saved story highlights, 'reels' = reels videos, 'all' = everything. Combine with comma: 'posts,reels'.
download_timeoutopt	INT	180060–36000	Maximum time in seconds for this run. Default 1800s (30 min). For huge galleries: 7200=2hr, 14400=4hr, 28800=8hr. TIP: If it times out, just run again - gallery-dl automatically resumes from where it left off using the download archive!
extra_optionsopt	STRING		Advanced gallery-dl options (one per line or space-separated): • --limit N: Max N files to download • --range 1-50: Download items 1-50 only • --no-download: Simulate without downloading • --write-metadata: Save per-file JSON metadata • --ugoira-conv: Convert Pixiv ugoira to video • -v: Verbose output for debugging • --sleep N: Wait N seconds between requests • --retries N: Retry failed downloads N times Folder organization examples: • Kemono by post: -o directory=["{service}","{user}","{id}_{title}"] • Flat by user: -o directory=["{category}","{user}"] See docs: https://github.com/mikf/gallery-dl/blob/master/docs/options.md
seedopt	INT	00–18446744073709550000	Random seed to force re-execution. Click 🎲 to randomize or change manually. Useful for resuming downloads after timeout - change the seed to run again without modifying other settings.

Outputs (4)

Name	Type	Description
output_dir	STRING	—
summary	STRING	—
download_count	INT	—
success	BOOLEAN	—