Google launches Gemini 4 Argon for Robotics: scores 87% on ERQA, teaches humanoids from video
GoogleDeepMind · x · 2026-10-14
- Google DeepMind released Gemini 4 Argon for Robotics, a general-purpose physical reasoning model that scores 87% on the Embodied Reasoning (ERQA) benchmark using pure reasoning, outperforming every other frontier model.
- Learning from video: Argon taught Apptronik's Apollo humanoid to play cornhole just by watching a video — mapping the physics, coding its own simulation, and sinking 7 throws in a row.
- Real2Sim: reconstructs physically calibrated simulation twins from robot camera feeds, egocentric human video, or hand-drawn sketches, compressing weeks of manual modeling into hours.
- Four key workflows with representative prompts: robot control (MuJoCo practice, zero-shot control, closed-loop steering of world models — demoed on a Boston Dynamics Spot), Real2Sim reconstruction, physical understanding (dexterity labeling, dense value functions), and research/self-improvement via simulation and hardware iteration.
More from Embodied
- Recova Trains Robots to Self-Recover in Simulation, Cutting Human Intervention From 87.5% to 0% — Lianhuiq · 2026-10-14
- NVIDIA's AlphaNIST Learns Robotic Assembly From Scratch With Only Binary Rewards, Hitting 98.2% Across 20 Tasks — abhishekunique7 · 2026-10-14
- EU privacy watchdog calls for smart glasses to clearly signal when recording — Polymarket · 2026-10-14
- How Argon drives Spot: real2sim skill verification in MuJoCo and map-free navigation — ki_ki_ki1 · 2026-10-14
- Agent Argon reads Spot's SDK docs itself, builds trash-sorting skill on Boston Dynamics robot — chris_j_paxton · 2026-10-14
- Meta researcher recruiting multimodal and robotics interns for next summer, no background needed — liuzhuang1234 · 2026-10-14