Turbo lora in sd cpp
#1
by urtuuuu - opened
In your readme you wrote: --width 640 --height 384 --video-frames 25 --steps 4 --cfg-scale 1.0
I wonder why 4 steps without lora?
I'm trying to find the best way to use 4 step lora in stable diffusion cpp, but outputs are not great enough, i wonder if anyone found a solution. (tried different turbo loras with different strengths, sampling, schedulers...)