Asmov-1 robot policy 85% trained with 0.007% eval error, open source next week
ericjang11 · x · 2026-11-02
Asmov-1 policy is 85% done training using Cartwheel cross-embodiment, control, and a scaling laws dataset, sitting at a 0.007% error rate on all eval tracks. Open source release expected next week, with the RoboParty RP1 policy also finished and included.
Related event: Asmov-1 Robot Policy Reaches 85% Training with 0.007% Error Rate(2 posts)→
More from Embodied
- DynaRobotics hotel towel-delivery robot spotted by guest in the wild — JasonMa2020 · 2026-11-02
- Curabotics inks national deal with Capstone, brings nurse robots to hundreds of US hospitals — MatthewChang · 2026-11-02
- Robotics needs its own CERN or Hubble to cut through the noise — DJiafei · 2026-11-02
- Dev adds an agent to a smart switch — talk to it via Morse code taps — davidcrawshaw · 2026-11-02
- Three-layer robotics stack — VLA world model plus RL whole-body controller — is already SOTA — zhaoran_wang · 2026-11-02
- CorebrumOS Physical AI OS now runs on NVIDIA Jetson with 8 agents — chrismatthieu · 2026-11-02