SYLIN

Stanford University

Maker · 2 entries

Everything from Stanford University

NameWhat to knowPrice or memoryRanks
UniActionModelOne video model controlled by forces, camera, hand or robot pose; built on Wan 2.2; code, weights, training scripts released
Next-Concept ModelingTechniqueStanford generates text coarse-to-fine, in parallel within each stage; up to 21x faster than diffusion LMs, 31x on summarization