requires to modify their pipeline which i dont want to
requires to modify their pipeline which i dont want to
Fbecause it would make it harder to merge once they improve
Ryea of course
Rby the way, your gui says v16 not v17
R
Flet me fix
Fplease do update
Ralso
![Dazzastrous [4090]](https://cdn.discordapp.com/avatars/871389071686135838/c4a5f4d81d9eeb18d856295d692e2c07.webp?size=40)
Dis that v18 now 
Fv17
Ri am getting OOO with an A6000 48gb. Seems like i still need the memory optimisation ??
R
R
R
Fyes i know
Ffor very high
Fyou need 80 gb a100
Roh
Ri didnt realise that
Rso the A6000 is not good, because same price as 4090 but slower chip
F
Fit is useful for vram
Ryea but but almost x2
Rwill it cut in half the compute time compared to 4090?
Fnope
Ffor single gpu
F4090 fastest
Funless you get like h100
Fthese are useful for high vram demand
Ryes but A100 no optimisation vs 4090 with optimisation, which one is faster, less time, by how much? Significant?
Fi didnt test
Fgive it a try
ROk, i cant find an h100 on runpod, only on community
Rand the bandwith is very low
Ryea there are no h100 or a100's avaialbe at all
Fye they are in super demand
DSupir Demand
Fstage 1
K@Dr. Furkan Gözükara seems to be broken, when ever i click apply stage 1 before stage 2, the final image is the blurry one
Rfound an a100
Rgonna test it
Fye stage 1 useless
Dwait so we only use stage 2 now?
![Dazzastrous [4090]](https://cdn.discordapp.com/avatars/871389071686135838/c4a5f4d81d9eeb18d856295d692e2c07.webp?size=16)
Fwe were always
Dso stage 2 becomes stage 1
Kalso maybe a button to stop the batch upscaling in newer version?
Fye
Flet me see if possible
Rdid you some testing between jugger9 and sdxl base?
Kyep i sent you a dm
