Replies: 3 comments 7 replies
-
hi,
Yes, I don't know if this is already a feature but I'll look out for it..
And finally thumbnails, we can add tons of tags, but nothing describes a model better than an example image, a nice model selector overlay with names, images, tags and trigger words would be awesome. This one is available as an extension for now, check https://github.com/AUTOMATIC1111/stable-diffusion-webui/wiki/Extensions#sd-model-preview |
Beta Was this translation helpful? Give feedback.
-
Just FYI under |
Beta Was this translation helpful? Give feedback.
-
Beta Was this translation helpful? Give feedback.
Uh oh!
There was an error while loading. Please reload this page.
-
I'm not sure if I'm suggesting these things in the right place, but I think there are enough like minded users around here that would deliver the message to the right ears.
With the growing amount of models and the considerable space they occupy on our hard drives, keeping multiple set of the same files for different applications would become an issue. It would be great if the developers add the ability to load models from an outside directory. The same for the embedding and everything else.
The ability to categorize the models into folders can also be very useful, for example in case of 1.5 vs 2.1 models knowing what version we're using through a folder structure can be very helpful. And maybe by the type of image they produce.
The ability to add a text files for additional info on the models would be great as well, perhaps a text file with the same name as the model, and when the model is loaded, the text in that text file would show in the webui. This text file can contain the essential trigger words that we need for some of models, and have to look for them every time we load them. Maybe some tags to categorize the models with. Perhaps these info would be useful when hovering a model before loading it.
And finally thumbnails, we can add tons of tags, but nothing describes a model better than an example image, a nice model selector overlay with names, images, tags and trigger words would be awesome.
What's your opinion on these improvements?
Beta Was this translation helpful? Give feedback.
All reactions