Show HN: We compressed a 31B video+audio diffusion model to 14B, 6 steps, INT8 Posted by jiant on Saturday November 07, 2026 @10:16AM from the show-hn dept. https://github.com/Video-Rebirth/hyperflow-h3-flash ( Read More... | no comments | on HN )
HyperFlow – one LoRA for all MiniMax-H3 tasks (ref2va/t2va/fl2va) Posted by jiant on Friday September 18, 2026 @12:37PM from the github.com dept. https://github.com/Video-Rebirth/hyperflow ( Read More... | no comments | 2 points | on HN )