#BasicText2Image
18 messages · Page 1 of 1 (latest)
Good job! I was looking for this couple days ago when I wanted to learn how nodes work and I ended up disassembling big graphs...
This should be added as a default "starting preset" in Invoke!
idk for sure if you have a low vram gpu it might be good to delete the progress image node. not sure if it uses a lot of vram or not but I have 6GB vram and it doesn't really slow things down so don't think it is a major problem
I'm not sure if low VRAM should be a thing to consider as Stable Diffusion is using Shared GPU memory - my RTX2060 is now 6+16GB. ^^
updated the graph, now added the ability to select the amount of images before generating using the size parameter from random range
I'm gonna try your graph today and did you have any problem with normal generation? or the setting you use I also have 6vram too (RTX2060)
not sure what you mean, this graph works for any system that can run invokeai
I mean, I'm sure your graph could work perfectly but instead of using graph the normal method I have problem with upscale I can not upscale with my 6vram
idk about that, haven't done much upscaling
by the way I use your graph and after generate image it's appear to be the grey noise image
Oh ok thank youu I'll try , so far most scheduler still not work LOL
what about eun or something?
I know the issue now is my model thank you for your help very much
I'm really surprised something like this wasn't in by default