This extension no longer functions for more modern releases of Automatic1111 for reasons that I, a layperson, don't fully understand. All I know is that it is still functional with v1.4.0 of A1111, and that versions from v1.6.0 onwards are not compatible with this extension.

The customizability of this extension was a big part of easing my learning curve for A1111, and to be 100% honest I'm still on v1.4.0 to today because moving beyond that version breaks compatibility (without this extension, you just get the poor-quality regular layout of A1111 in future versions).

I'm guessing this extension has been abandoned?

Settings:

Resulting view:

Version:

Well, I think this extension is outdated. It is for a version of the webui without the tabs "Generation", "Textual Inversion" and so on ...

I would like the principles behind the idea, but first the layout of the webui has to be changed again. The developers should use the regions as drop-zones for the present elements and allow the users, to define where they want these elements.
In some parts (User Interface settings) its already working like that, but unfortunately it's not possible to split the lower window into separate columns, where one could have the prompts to above the tabs, but in the same column.
Also to design the upper tabs row as icons or with vertical text and arrange it as sidebar to the left, like InvokeAI, would be nice.

https://www.reddit.com/r/StableDiffusion/comments/12b45ys/comment/jg33j5p/?context=3
Continue the topic here.

OK. To maximize the image, I think the first thing is to be able to scale the image.
Currently, the generated image will only be displayed at its original size. However, for large screens or people who have difficulty generating large images, the output at its original size is still not large enough.
(Because what we want is for the generated image to be maximized when displayed, so we can rule out clicking it again to enlarge it to full screen.)
For me, the ideal situation would be to allow the output image to automatically scale to fit the image frame.

Furthermore, as the number of controls on the left side increases, users need to scroll and click more frequently.
However, at the same time, the position of the image frame on the right side will also move, causing users to scroll repeatedly to view the image.
If the image frame can be fixed on the right side and not affected by scrolling, this problem can be solved.
The generate button should also be fixed, for example, on the outer edge of the upper left corner of the image frame. For users who prefer to click the generate button rather than use CTRL+ENTER, they can click the generate button directly no matter where they scroll to.

Next, I think we can move all the buttons above and below the image frame to the left, so that the image frame can be fully expanded to the top and bottom edges of the window. The remaining task is to optimize the arrangement of elements.

For some elements, I think we can consider not moving them to the left, but hiding them temporarily until the mouse hovers over them, or providing a completely hidden option. For example:




I think these are suitable for temporary hiding.

And this can be completely hidden. Users don't need to check it often, so they don't need to let it take up space all the time.