Cinematic World Building using H3 r2v: 'Through the Sands' visual demo
R34vspec · reddit · 2026-08-24
The author shared a video titled 'Through the Sands', generated using the H3 (r2v) model, demonstrating its capabilities in cinematic world building and visual storytelling. The generated shots were described as impressively realistic, even giving the author goosebumps upon first viewing. This is the second part of a series, with the finale planned to complete the project.
More from Multimodal
- 4DAnyone open source: Turn casual phone video into consistent multi-view 4D avatars — aigclink · 2026-08-24
- Zhejiang University & Ant Group open source 4DAnyone for high-quality 4D avatars from phone video — aigclink · 2026-08-24
- MiniMax-H3-Fun-Controlnet-Union: Single Checkpoint for 5 Control Modes & Inpainting — GreyScope · 2026-08-24
- WAN 3.0 generates 30s video in one take, challenges SeeDance 2.5 — mark_k · 2026-08-24
- SenseTime SenseNova-Vision: Unifies Visual Tasks via Multimodal Generation — rsasaki0109 · 2026-08-24
- AI-generated anime football highlights look incredibly realistic — technollama · 2026-08-24