Version 1.3.8 du 12/08/2025
https://github.com/CyberTimon/RapidRAW/releases/tag/v1.3.8

Improvements :
Even more bug fixes following up on yesterday
Added a duplicate mask context menu shortcut
The image canvas now correctly updates the preview if just the rotation or the aspect ratio has changed
RGB curves are working properly now when they have just 2 points (no added point)
Added support for sony a6400a
AI subject masks work again (they were buggy yesterday because of the new quick eraser tool)

Version 1.3.9 du 17/08/2025
https://github.com/CyberTimon/RapidRAW/releases/tag/v1.3.9

AI Sky Masks
RapidRAW now supports one click sky masks. Just hit the sky mask button and RapidRAW will take care of masking the sky for you.
The pretrained model I’m using comes from this repo and shares the same architecture as the AI foreground model (u2net).

Optimised Image Editing
One of the biggest bottlenecks in RapidRAW used to be the slow base64 bridge between the backend and frontend. Encoding/decoding large images added a ton of overhead when applying adjustments or just opening files.

That’s now fixed. I switched over to Tauri’s IPC response, which avoids the whole base64 step. The result: editing feels way smoother, and fullscreen zoom actually loads the full res image so you can check every pixel.

je met de côté si nécessaire ça m'a l'air pas mal merci

Sous le coude, donc

Citation :
Tu n'avais pas vu l'info dès le début du fil ?

Faut croire que non. Tu pose une question dont on connaît déjà la réponse.
Quelle perte de temps !

Citation :
C'est comme "sous le coude" qui ne fait que reprendre ce qu'a dit Keyra dans le post précédent.

OK j’aurais du faire un MP à keyra74, car il est toujours intéressant de savoir comment sont perçus les fils de nouveaux logiciels.
Il est très rare qu'il y ait des retours sur ce type de fil.

Version 1.3.10 du 25/08/2025
https://github.com/CyberTimon/RapidRAW/releases/tag/v1.3.10

You can now apply LUTs directly to your images, with support for .cube, .3dl, .png, .jpg, .jpeg, and .tiff files. This makes it easy to add color grading or use a LUT as a creative starting point.

Version 1.3.11 du 01/09/2025
https://github.com/CyberTimon/RapidRAW/releases/tag/v1.3.11

ComfyUI Integration :
ComfyUI integration has been significantly improved to make it more accessible and user-friendly. You can now:
Select a custom inpainting workflow or use the built-in default workflow.
Define and map which models should be loaded.
See which custom nodes you need to install, in order to use the default workflow.
RapidRAW automatically handles downscaling, mask generation, patch layer composition, and saving. These enhancements make ComfyUI integration practical for everyday use.

Note : Some bugs and unfinished elements may still be present. Please report any issues you encounter. Also, if you have a better & easier inpainting workflow, please share it with me as I'm not satisfied with the current one.

Chromatic Aberration Correction :
A new tool for chromatic aberration correction has been added.
image

Other Improvements :
Changing folders now cancels previous thumbnail generation and folder indexing jobs, preventing performance slowdowns on many machines.
LUT's parser is more robust / less error prone now.
Mask grow / feather parameter now correctly applies on the final image.