Hello everyone,
I am aware that the Stable Diffusion model for use with the Diffusers library is available both in full precision and half precision (fp16) versions, and that using the half precision version saves VRAM.
However, are there any other practical differences? Is the half precision model slower, or do the generated images tend to have less quality? When would you want to choose the full precision model over the half precision one?
Thanks!
Hey, been a while since you posted and to be honest I hope you found the answer before now. However, on the off-chance you haven’t found an answer yet, yes, the half precision (fp16) is slower but in my opinion that’s not that big of a deal since the results you get are almost, or entirely the same. To answer your last question, stable diffusion version 2.1 requires full precision and I don’t know if they plan on making half precision possible sometime in the future. Hope this helps!