fractal-fumbler

Results 36 comments of fractal-fumbler

[Bug]: 20% reduction in performance for single image generation due to latest commit fix for VRAM spike

> What anybody thinks? tried this one and didn't have spikes for batch size=1 or batch size=2 with generating images with resolution 768x768 ```python actual_batch_size = samples_ddim.size(0) if actual_batch_size>1: x_samples_ddim...

[Bug]: 20% reduction in performance for single image generation due to latest commit fix for VRAM spike

> with resolution 768x768 with resolution higher than 768x768 lag spikes appearing again for a single image generation (for example, 832x832)

Add upcast options, full precision sampling from float16 UNet and upcasting attention for inference using SD 2.1 models without --no-half

@ddvarpdd is your model for SD-1.5 or SD-2.0 or SD-2.1?

Add upcast options, full precision sampling from float16 UNet and upcasting attention for inference using SD 2.1 models without --no-half

so, @brkirch, 96093475731c2f95fba3911ee66e5065deb21005 working on SD-1.5 and SD-2.0 (for me on pytorch 1.13.1), including embeddings :) also tested on 280083c9c30058d092c6b6f6aadac5e669b322fc - works getting increased speed with Euler a from 1.3...

Add upcast options, full precision sampling from float16 UNet and upcasting attention for inference using SD 2.1 models without --no-half

so amd (rocm) problem with SD-2.1 is that there is no xformers available and amd users can't make SD-2.1 to force to use fp16 with, since it is only working...

Add upcast options, full precision sampling from float16 UNet and upcasting attention for inference using SD 2.1 models without --no-half

after pulling latest commits with 96093475731c2f95fba3911ee66e5065deb21005 i am getting vram is around 4.7-4.8 GB with generating 768x768 and with 1024x768 (hires fix) upd: w/o medvram it's like 5.0-5.1 GB VRAM...