leszekhanusz / diffusion-ui

Frontend for deeplearning Image generation
https://diffusionui.readthedocs.io
MIT License
142 stars 12 forks source link

[Enhancement] Allow the Changing and Reloading of Hypernetworks and Models of the Webui Backend #66

Open LumiWasTaken opened 1 year ago

LumiWasTaken commented 1 year ago

It would be very neat to have a quick access to change the current model and or hypernetwork

This would open up the tool to more experimentation for many. Testing how the inpaint behaves differently between multiple Models

Bonus: Having the same Seed grouped with different model-hashes for easy comparison later on.

leszekhanusz commented 1 year ago

T̶h̶a̶t̶ ̶w̶o̶u̶l̶d̶ ̶b̶e̶ ̶n̶i̶c̶e̶.̶ ̶I̶s̶ ̶i̶t̶ ̶a̶l̶r̶e̶a̶d̶y̶ ̶s̶u̶p̶p̶o̶r̶t̶e̶d̶ ̶o̶n̶ ̶a̶u̶t̶o̶m̶a̶t̶i̶c̶1̶1̶1̶1̶?̶ Yes!

LumiWasTaken commented 1 year ago

T̶h̶a̶t̶ ̶w̶o̶u̶l̶d̶ ̶b̶e̶ ̶n̶i̶c̶e̶.̶ ̶I̶s̶ ̶i̶t̶ ̶a̶l̶r̶e̶a̶d̶y̶ ̶s̶u̶p̶p̶o̶r̶t̶e̶d̶ ̶o̶n̶ ̶a̶u̶t̶o̶m̶a̶t̶i̶c̶1̶1̶1̶1̶?̶ Yes! I am not sure if ur being sarcastic.

But i would like to immerse myself into one tab rather than having to jump between the automatic1111 Ui and Yours, so changing the model and Hypernetwork from there, where the settings are already feeling much more responsive (as in Ux) would be a big plus.

leszekhanusz commented 1 year ago

Not I was not sarcastic :smile: I edited my comment. I really did not know.

leszekhanusz commented 1 year ago

I'll try to implement it.

LumiWasTaken commented 1 year ago

I'll try to implement it.

Lovely <3

leszekhanusz commented 1 year ago

It should work now (#67)

leszekhanusz commented 1 year ago

How to easily compare multiple models:

LumiWasTaken commented 1 year ago

It should work now (#67)

Would you mind adding Support for Hypernetworks aswell? I also have the issue, when i merge models, i do not have a way to let the diffusionui know that it should re-read the model list. After generation it does not show up there.