FastVideo/FastVideo-FastH3-Trim-Comfy · an experimental pruned FastH3 model: 42 transformer blocks instead of 50 and a rank-16 timestep conditioning, sampled in 8 steps. It is 4.2× smaller than base

Submitted ⁨⁨1⁩ ⁨week⁩ ago⁩ by ⁨Even_Adder@lemmy.dbzer0.com⁩ to ⁨stable_diffusion@lemmy.dbzer0.com⁩

https://huggingface.co/FastVideo/FastVideo-FastH3-Trim-Comfy