FLASH: GPU-native deformable sim learns cloth folding in minutes on one RTX 5090
ChongZzZhang · x · 2026-10-26
The FLASH paper (arXiv) introduces a GPU-native simulation framework for contact-rich deformable manipulation, rebuilt from scratch around an NCP solver for fine-grained GPU parallelism. It simulates 3M+ degrees of freedom at 30 FPS on a single RTX 5090, and policies trained purely on FLASH synthetic data in minutes achieve robust zero-shot sim-to-real transfer, validated on real robots doing cloth folding. The fuller FBA solver is dropping soon.
Related event: Deformable Object Simulator FBA Set for Release, Joining NVIDIA Newton(3 posts)→
More from Embodied
- Folding-screen hinges: dozens of ultra-precise parts, one grain of sand can kill the panel — BenBajarin · 2026-10-27
- No π0 weights changed: Motoniq layer lifts real-robot precision fitting from 0/10 to 9/10 — hbouammar · 2026-10-27
- $2,000 iPhone Duo already showing creases and cracks two days after launch — 404 Media · 2026-10-27
- Chums: open-source Apple Watch app puts Claude Code and Codex on your wrist — brandon_galang · 2026-10-27
- Dev uses Grok bot to turn ESP into an iPhone accessory for syncing robot servos — Baconbrix · 2026-10-27
- NVIDIA releases MotionBricks: real-time humanoid motion generation at 15,000 FPS with 2ms latency — _akhaliq · 2026-10-26