4 ms·
Nvidia’s New StyleGAN-T Is 30X Faster Than Stable Diffusion
- brucethemoose2 4y agoRelease aimed for the "end of March" https://github.com/autonomousvision/stylegan-t/issues/3 https://github.com/autonomousvision/stylegan-t/issues/3 I took a crack at hacking temporal stability into huggingface sd, but (if it has image2image) this is much more promising since its so fast.