Hey guys, did you take a look at: https://github.com/VoltaML/voltaML-fast-stable-diffusion Seems to be a valid way for speeding up generation by a lot (3x faster)... it optimizes the model using TensorRT for nvidia GPUs.
Hey guys, did you take a look at:
https://github.com/VoltaML/voltaML-fast-stable-diffusion
Seems to be a valid way for speeding up generation by a lot (3x faster)... it optimizes the model using TensorRT for nvidia GPUs.