China Mobile open-sources Open-RAIL, a real-time execution layer for VLA models and humanoids
CyberRobooo · x · 2026-09-23
- China Mobile released Open-RAIL, an open-source real-time asynchronous inference middleware bridging VLA and World Action Models with humanoid robots, addressing the mismatch between chunked model outputs and the need for fast, continuous motion.
- It handles asynchronous inference, temporal alignment, and trajectory smoothing, claiming support for 20+ VLA/WAM models and 4 humanoid types (Unitree G1, AgiBot G1, China Mobile Lingxi, NAVIAI-WA2), plus real-robot data collection and human-intervention tools.
- In real-robot tests it achieved up to 2.09× faster task execution while reducing motion jitter, easing model portability across robot bodies.
More from Embodied
- Homebrew robot arm build log Day 7: geartrain CAD, wiring and gripper sizing — _Stocko_ · 2026-11-11
- Perceptron AI launches EgoHome, an open benchmark for sub-second egocentric hand understanding — rohanpaul_ai · 2026-11-11
- Runway open-sources Praxis-1, turning video pretraining into real robot control — runwayml · 2026-11-11
- Unitree G1 humanoid wearing VR goggles pilots an A1 drone in land-air robot demo — CyberRobooo · 2026-11-11
- CoRL Spotlight Meta-Ctrl predicts robot plan failures, Llama 3 8B beats GPT-5.5 on VirtualHome — guyvdb · 2026-11-11
- Luna, a jewelry-style AI wearable that speaks before you ask, launches on Kickstarter — nikola_mr64990 · 2026-11-11