Have't done quantization or max autotune yet, getting 30FPS just from basic compile on RTX 5070 on a laptop :D Next step is to get fp8 60fps, then after that I gotta plug the upsampler in. These models will run on everyones hardware!

Aug 27, 2025 · 8:48 PM UTC

5
3
1
29
0
Also wanted to add there are 3 forward passes per frame, and VAE+rendering is adding overhead, so this is actually ~120 bf16 forward passes per second.
1