v1.14.0

• Allow setting directory containing saved auto-captioning models (File -> Settings) (#46)

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.

v1.13.0

• Add context menu action to invert selected images in the image list (shortcut: Ctrl+I)

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.

v1.12.2

• Set trust_remote_code=True when loading auto-captioner processors (#45)

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.

v1.12.1

• Allow loading auto-captioning models from local paths by typing path in Model box

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.

v1.12.0

• Add Include in caption parameter for auto-captioning
  • Allow specifying words or phrases that should appear in generated captions (see the README for details)
• Maintain word wrap when editing multiline tags (#38)
• Exclude JSON files when loading directory

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.

v1.11.2

• Fix a bug where CogVLM would raise an error when captioning a non-RGB image

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.

v1.11.1

• Disable scrolling of some widgets when they are not in focus (#35)
• Improve default prompts for CogVLM and LLaVA auto-captioning models
• Fix a bug that would sometimes prevent the program from starting

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.

v1.11.0

• Add support for CogVLM auto-captioning model (#32)

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.

v1.10.0

• Add support for InstructBLIP and KOSMOS-2 auto-captioning models

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.

v1.9.1

• Support loading auto-captioning models in 4-bit on Windows (#29)

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.