ForgeWM: Progressive Training for Video World Models
ForgeWM · hf · 2026-08-21
ForgeWM proposes progressive causal training to distill bidirectional video generators into efficient few-step interactive world models. It supports aligned discrete and continuous controls, enabling low-latency interaction and replay-time refinement.
More from Models
- "Cola Ox" model launches free beta, claiming Terra-level ability at Flash speed — oran_ge · 2026-08-21
- Ox Alpha Review: Faster than Kimi K3, Rivals Claude — teortaxesTex · 2026-08-21
- OpenRouter's 'ox-alpha' model identified as GLM-5.3 — realmrfakename · 2026-08-21
- Liquid AI Releases DSpark Draft Models for Up to 3.18x Faster Decoding — JosephJacks_ · 2026-08-21
- User Notes Claude Output Needs No Watermark Due to Distinct Style — andrew_n_carr · 2026-08-21
- OpenCode Offers Free Access to Ox Alpha: 1M Context and Multimodal — TJLarkin23 · 2026-08-21