Comment by naillo

Might be that there's some degredation but I think it's pretty close. Anyway I'm using their 'official' fp16 version which they might be doing some extra magic on idk. I.e. via

  StableDiffusionPipeline.from_pretrained("CompVis/stable-diffusion-v1-4",
    revision="fp16", torch_dtype=torch.float16, ...)