0.7.0

What's New:

• X/Y/Z Script support (4 available axes: CodeFormer Weight, Restorer Visibility, Face Mask Correction, Face Models)
• You can now blend faces to build blended face models ("Tools->Face Models->Blend") - due to popular demand
• CUDA 12 Support in the Installer script
• New tab "Detection" with "Threshold" and "Max Faces" parameters
• RGBA support
• Different fixes and improvements

0.6.1

What's New:

• 'Save original' option works fine now when you select 'Multiple Images' or 'Source Folder'
• Random Mode for 'Source Folder'
• 'Force Upscale' option inside the 'Upscale' tab: ReActor will run the Upscaler even if there's no face is detected
• ReActor shows filenames of source images in-process when the multiple images mode or the folder mode (random as well) is selected
• No saved images bug is fixed #342 (when selected one source image and multiple images)
• Different fixes and improvements

0.6.0

What's New:

• Multiple Source Images Support (you can load several images or specify a path to the folder containing face images)
• New Logo
• Adaptation to A1111 1.7.0 (appropriate GFPGAN loader + new enable feature in UI) + SD.Next UI-enable fix
• New URL for the main model file
• Different fixes and improvements

0.5.1

What's New:

• You can now save face models as "safetensors" files (stored in <sd-web-ui-folder>\models\reactor\faces) and load them into ReActor, keeping super lightweight face models of the faces you use
• "Face Mask Correction" option - if you encounter some pixelation around face contours, this option will be useful
• Fixes and improvements

Thanks @jiveabillion for adapting "Face Mask" solution taken from another excellent extension ototadana/sd-face-editor (thanks to @ototadana)

0.5.0

What's New:

• ReActor is now available in Extras Tab
• CUDA Support: full support of ORT-GPU alongside with ORT
• New option in Settings: Execution Provider selector (if CUDA is available), you can switch to CPU Execution Provider if you need to reduce VRAM usage
• Analyzer logic is updated
• A lot of fixes and improvements

Thanks @w-e-w for the "Extras Tab" solution and pointing the right direction with CUDA

0.4.3

What's New:

• A lot of fixes and improvements
• Source and Target images hashing -> good speed boost of images processing

0.4.2

What's New:

• Fixes and improvements
• ORT 1.16.0 fix (downgrading to 1.15.1, awaiting for the 1.16.1 patch)
• ReActor Force with NVIDIA GPU support is released
• CodeFormer weight slider is added
• Huge boost when swapping several faces, thanks @Dutch77 for spotting #95 and for showing the way of resolving it

0.4.1

What's New:

• UI rework
• "Save Original" option
• More unique filenames to avoid conflicts with other extensions
• Different fixes and improvements

0.4.0

What's New:

• ReActor external API
• From now on all the models (insightface buffalo and inswapper) are in the "models/insightface" directory (thanks @h3rmit-git for the idea and clarification)
• FIX: Extension can "interrupt" a process now
• FIX: New model link for inswapper model file
• FIX: WebUI API gender default condition when no gender parameters are passed via API (thanks @andypotato for spotting)

0.3.1

What's New:

• Gender detection added
• Little speed up when ReActor is not enabled (thanks @huaizong for noticing)