Releases: jhc13/taggui
v1.23.1
- Reduce unnecessary messages in the console during auto-captioning
The bundle is compressed as a .7z
file to stay under GitHub's 2 GB file size limit. You may need to install 7-Zip on your system to extract the files. Download the file for your operating system, extract it, then run the executable file inside.
v1.23.0
- Allow selecting multiple tags in the All Tags list for renaming or deletion (#118)
- Add support for the LLaVA-Llama-3-8B auto-captioning model (#127)
- Allow using the numpad Enter key when adding or renaming tags (#125 @yggdrasil75)
- Handle error when checking Exif tags (#123)
The bundle is compressed as a .7z
file to stay under GitHub's 2 GB file size limit. You may need to install 7-Zip on your system to extract the files. Download the file for your operating system, extract it, then run the executable file inside.
v1.22.0
- Add setting for file formats shown in the image list (#121 @yggdrasil75)
The bundle is compressed as a .7z
file to stay under GitHub's 2 GB file size limit. You may need to install 7-Zip on your system to extract the files. Download the file for your operating system, extract it, then run the executable file inside.
v1.21.0
- Allow selecting which GPU to use for auto-captioning on multi-GPU setups (in advanced settings) (#112)
The bundle is compressed as a .7z
file to stay under GitHub's 2 GB file size limit. You may need to install 7-Zip on your system to extract the files. Download the file for your operating system, extract it, then run the executable file inside.
v1.20.3
- Handle case where no tags are generated from a WD Tagger model without raising an error (#109)
The bundle is compressed as a .7z
file to stay under GitHub's 2 GB file size limit. You may need to install 7-Zip on your system to extract the files. Download the file for your operating system, extract it, then run the executable file inside.
v1.20.2
- Pin the version of the moondream2 auto-captioning model to the last stable version (#104)
The bundle is compressed as a .7z
file to stay under GitHub's 2 GB file size limit. You may need to install 7-Zip on your system to extract the files. Download the file for your operating system, extract it, then run the executable file inside.
v1.20.1
v1.20.0
- New features in
Edit
->Batch Reorder Tags
(Ctrl
+B
) - Allow sorting the All Tags list by frequency, name, or length (#40)
- Allow setting the click action in the All Tags list to add the tag to selected images instead of filtering the image list (#95)
- Support
*
and?
wildcards in the Images list and All Tags list filters (#24) - Improve the experience of editing, deleting, and reordering tags in the Image Tags list
- Get the list of auto-captioning models in the models directory faster
The bundle is compressed as a .7z
file to stay under GitHub's 2 GB file size limit. You may need to install 7-Zip on your system to extract the files. Download the file for your operating system, extract it, then run the executable file inside.
v1.19.0
- Add support for WD Tagger auto-tagging models (#76)
- Add support for LLaVA-NeXT (LLaVA-1.6) auto-captioning models (#50)
- Immediately display generated tags as separate tags without having to reload the directory (#88)
- Find and Replace (
Ctrl
+R
) improvements- Add option to find and replace in selection only (#83)
- Persist find and replace texts when the dialog is closed
The bundle is compressed as a .7z
file to stay under GitHub's 2 GB file size limit. You may need to install 7-Zip on your system to extract the files. Download the file for your operating system, extract it, then run the executable file inside.
v1.18.0
- Add
Discourage from caption
parameter for auto-captioning that prevents certain words or phrases from being generated (see the README for details) - Show
Cancel Auto-Captioning
button while generating captions to allow canceling an auto-captioning run
The bundle is compressed as a .7z
file to stay under GitHub's 2 GB file size limit. You may need to install 7-Zip on your system to extract the files. Download the file for your operating system, extract it, then run the executable file inside.