These tag list(s) contain tags + their aliases (active & deleted + chars/artists/franchise tags) from Danbooru and/or e621 and are for use with sd-webui-tagcomplete extension and ComfyUI-Custom-Scripts autocomplete feature (or AutocompletePlus / comfy-ex-tagcomplete) as well as Krita AI Diffusion, Stability Matrix and SwarmUI (and lots more probably).
There are 3 main versions.
Danbooru - Danbooru tags only (best for Illustrious and NoobAI/Pony if not using e621 knowledge)
e621 - e621 tags only (best for e621 models/usage only; e.g: NoobAI)
danbooru_e621_merged - a merge of both Danbooru and e621 tags (best for NoobAI/Pony based models especially if you plan on using e621 stuff and your extension doesn't support loading multiple tag lists at once)
There's also 🖌️ Krita AI Diffusion compatible tag lists inside the corresponding folder.
There is a 🇨🇳 Chinese version made by @UrielGnoyzaki (comment).
The a1111-sd-webui-tagcomplete extension already comes with the 22.12.2024 tag lists so you don't need to download those from here.
22.12.2024 is the recommended version for overall use case if you don't want to switch a lot. "01.09.2025 ChenkinNoob 01" and similare are release made for the model of the same name, but not exclusive to, just use them for what you see fit. The ones that are have a more recent date are mostly there for future proofing and the older ones usually have worse filtering/merging but the site-indivdual tag lists should be ok to use.
You can find the full archive of all tag lists (including other misc stuff like some raw data) that have been made on the archive GitHub repository. (auto updates every months as GitHub actions artifacts, and uploads to repo every 3 months)
The script(s) with which I made these tag lists is open source on my GitHub.
Some info on the install process for some tools can be found on the main page of the repo (readme.md)
Some rather trivial info:
NoobAI 1.1/1.0 Eps/Vpred knowledge cutoff for danbooru is roughly before 23rd October 2024
NoobAI's e621 dataset is only up to date until early 2024 (Yi from Nine Sols will won't work for example) meaning older tag lists for e621 specifically might feel better to use
Illustrious v0.1/1.0-2.0 danbooru knowledge cutoff is roughly june 2024
RouWei has a slightly more up to date dataset (april 2025 for 0.8, dec 2024 for 0.7 or so) for those who use it and see this
Generally I don't think much has changed over the few months so the 22nd december 2024 will still be the recommended version.
The Danbooru tag scraping part was originally borrowed from BetaDoggo over here (many thanks)
Also a thank you to e621 for having every list easily accessible
also pandas dataframes are kinda cool.
Description
Created on 1st December 2024 (Upd. reason since it's out of schedule: NoobAI EPS 1.1 release)
Minimum post count threshold is 40 (-10 from last version since noobai seems to know more than expected)
Characters/Artist/franchine names + Active and deleted alias tags are included (pending alias tags from e621 are excluded)
Next update will include a major script update and remove useless tags like conditional_dnp and banned_artist etc.
If there are any performance problem with the huge combined list (even though there shouldn't be any) please tell me in the comments. I'm sure there's things that can be optimized for the list, I already got some rather trivial things in mind. I might also lower the post count threshold if noobai knows those tags too and there aren't any performance problems.
After this version there will be some changes to the tag lists and script mainly when I got time:
deleted alias tags will be ignored if they exist as a full tag (don't think thats handled yet or is a real issue anyway)
slight changes to the file names to make it easier to differentiate between different versions of self made tag lists using the script on GitHub
As always, you can find the tag lists on GitHub.
If you are using just the E621 tag list I recommend using the oldest version of it since there doesn't seem to be a model trained on up to date e621 dataset
I will also probably do some work to be able to create artist wildcards based on post threshold
FAQ
Comments (23)
Thanks to the author for the merge, but I need to mention that if you are using tagutocomplete and frequently use the '?' function in front of the candidate words to query the meaning of the tag, you must rename the downloaded file to danbooru or e621, otherwise the '?' will not appear in front of the tag.
Edit: The new version of tag autocomplete has been corrected. Please see the comments for details.
Thanks for the comment
I wasn't aware of the feature but I have now uploaded and replaced the zip file with the same tag lists in it but the e621/danbooru only tag lists are now prefixed with tagcomplete-wiki-links compatible names (danbooru and e621) to work out of the box with that feature without renaming the files manually.
For the merged tag list I might try to explore ways to get the feature to work with it, possibly through editing the tagautocomplete extension code and making a merged tag list with an extra column to signalize which booru site a tag is from
This would be an option for people who want to put in the extra work to modify the extension code (pasting in what i modify) or maybe clone a fork of the extension with merged list wiki link support (I thought about making a pull request/contribution to the original extension repo but i don't think it'd be good/accepted since the merged list isn't really a standard and has alias tags conflicts)
Just wanted to give you a quick update: the extension now comes with the 22.12 version of the tag lists by default including the merged one, and it works correctly with the wiki links feature
Does it include characters?
Yes,
general tags, characters, artist names, franchise/copyright tags and meta tags should be included
if you want a custom version i can make one.
@DraconicDragon Awesome. I think it would be very helpful if you can separate characters in groups, like male female etc
@Rating_Agent Hmm, i don't think that will be possible, at least not to my knowledge. I don't think Danbooru or e621 save gender information or similar with a character tag
The best I can do right now would be creating lists without aliases (or without deleted/pending ones),
lower post count threshold for the lists,
exclude specific tag categories a.k.a. creating an artist, character and/or franchise only list + having it based on post count (to be used as wildcard, which i'm already planning to do)
possibly more stuff related to editing the current lists, but thats what i remember at the top of my head
@DraconicDragon there is a gender tags,1boy solo and 1girl solo
@Rating_Agent yeah i thought about that over the night, maybe i'll give it a try after trying to get tagautocomplete wiki links working with the merged tag list
haha update after 25 days
i currently have a prototype-kind of script running for this right now and testing it
it runs at about 1 second per character tag IF the gender can be determined using the first page of posts results so chars where the gender isn't clear like konpaku_youmu_(ghost) or hilichurl_(genshin_impact) will take substantially longer
I've tried to keep false positives as low as possible and in second-worst case they should be filtered out as indeterminate like hilichurl_(genshin_impact) since its almost impossible to determine the gender for that and it'd need a second pass because its almost all sex scenes with more than 1 person or its humanization and then its a female one
I haven't tested low post count characters a lot (>10 posts) but its possible many get filtered out as indeterminate if they don't have any posts with tags that fit my filters
@DraconicDragon Wow, nice!
@Rating_Agent
https://civitai.com/models/1087743
theres some stuff still to be sorted out (see description known issues) but generally almost all humans should be correctly sorted
CSV with the tags combined sorted by post count like the text files but with post count column https://github.com/DraconicDragon/dbr-e621-lists-archive/blob/main/wildcard/2024_12_22/
edit: just found this https://huggingface.co/datasets/deepghs/character_index
it only goes down to chars with ~300 images apparently, i dont know the specifics
ill probably make a version based on this and my own
it also has gender approximation which is likely done through some model looking at the jsons while i just scraped danbooru posts for each character tag and filtered them out
awesome
Thanks for creating this and keeping it up to date. You're awesome o(*￣▽￣*)ブ
im sorry im new but how do i install it on comfy ui?
You need to install pythongosssss/ComfyUI-Custom-Scripts (best through ComfyUI Manager)
After you installed that, restart comfy and then go to the Settings -> search for "autocomplete"
If custom-scripts was installed correctly, there should be a setting called "Auto Completer" with a button saying "Manage Custom Words"
Click that button and wait for a popup/window with the title "Custom Autocomplete Words" to appear.
Easiest way is probably to just paste the contents of one of the CSV files you want to use into the text area, thats what I did.
(It's also possible to use a link like this one https://raw.githubusercontent.com/DraconicDragon/dbr-e621-lists-archive/refs/heads/main/tag-lists/danbooru/DBR_tags_2024_11_30.csv (danbooru list) from my github repo but I think in your case its easier to copypaste the contents into the text area)
@DraconicDragon Is it possible to install that without Comfy Manager? Like, manually? Anytime I try to clone using command prompt it sends it somewhere my Comfy doesn't recognize and I have to hunt it down and move it, for some reason. It's a pain, so I prefer to just manually download and drop the dependencies where they need to go if I can.
@Anistasiata ComfyUI manager should just run the git clone repo command inside the into the ComfyUI/custom_nodes folder, so yes, it's entirely possible to install without comfy manager, it's just that it's easiest with it since it eliminates the need to use the command line.
Either you do it like it says on the pythongosssss/ComfyUI-Custom-Scripts repo (it has an "Installation" section at the top) and run the git clone command inside the custom_nodes folder...
Or you download the repo as a zip (using the green button labelled "Code" on GitHub), unpack the zip and put the folder you got from the zip inside the ComfyUI/custom_nodes folder
I haven't tried the latter method but if ComfyUI works how I think it does, it should work that way
@DraconicDragon Thanks it work!!!
If i'm running and use reforge or forge can and do I need this script or not ?
There's no script to use (i do provide one but I'm getting to that later).
This page provides CSV files which include e621/danbooru tags to be used in either of the in the description mentioned tag-autocomplete extensions.
I do provide a python script to create these CSV files which is on my GitHub, but unless you want to create them yourself you don't need any script from me, just the CSV files in the zip you can download on this CivitAI page and an extension that can work with them.
This looks better than the default tags because it shows the danbooru tag corrections, such as "sitting_on_knees" showing that danbooru uses "seiza" instead. Amazing, no more needing to jump between Danbooru and my workflow to get the tag correct! Thank you so much!
可以