new | ask | show | jobs
Show HN: We compressed a 31B video+audio diffusion model to 14B, 6 steps, INT8
4 points by jiant 2 hours ago | 1 comments
jiant 2 hours ago  [ - ]

[flagged]