Argon Robotics Trains Robot Policy 4x Faster Than Teleop Data, 95%+ Success Over 500 Real Runs
chris_j_paxton · x · 2026-09-25
Argon Robotics published its first research blog, claiming it trained a robot policy that runs at roughly 4x the speed of the teleoperation data it was trained on, while holding a >95% success rate across 500 real-robot runs.
The company positions itself as building "the fastest autonomous robots in the world," arguing that for three centuries machines have outrun humans but execution was always throttled to human wrist speed — and that speed is the final frontier to unlock before robots can assist humans in the real world.
Argon also lists 3 open roles and stresses a high, non-moving hiring bar.
More from Embodied
- NYU's Steertuning Lets a Robot Learn a Chore From 5 Minutes of Human Demo — CyberRobooo · 2026-11-11
- Boston Dynamics Atlas demoed with Blender 5.3 — TomLikesRobots · 2026-11-11
- Lamport Robotics posts robot 'wake-up test' demo video — R4nsack · 2026-11-11
- Embodied AI Faces Two-Order-of-Magnitude Data Gap as Industry Heads to 10M Hours — GaryMarcus · 2026-11-11
- NVIDIA hosts CoRL 2026 meetup on scaling robot training with Isaac Lab — m_wulfmeier · 2026-11-11
- Lighthouse extends Obsidian welding foundation model to run on any robotic platform — bennash · 2026-11-11