use the --vae_tiling args seems using absurdly high VRAM when vae decoding, >2x than not using it, is there something wrong?
non-square config seems generation at a speed near bigger one, rather than between (2048x1024 as slow as 2048x2048, not in between 1024x1024 and 2048x2048), is it intentional?
--vae_tiling
args seems using absurdly high VRAM when vae decoding, >2x than not using it, is there something wrong?