Google's Gemini 4 Argon for Robotics scores 87% on ERQA, tops all frontier models
mihdalal · x · 2026-10-14
Google announced Gemini 4 Argon for Robotics, claiming a step change in physical understanding, code generation and visual-semantic reasoning. Using pure reasoning it scores 87% on the Embodied Reasoning (ERQA) benchmark, outperforming every other frontier model.
The post details four application areas with prompts:
- Robot control: zero-shot control of a Boston Dynamics Spot (12-DoF legs + 6-DoF arm) learning to step on a foot-pedal trash bin — Argon builds a MuJoCo digital twin to practice before executing on hardware; also closed-loop steering of video world models.
- Real2Sim: reconstructing physically calibrated simulation twins from robot camera feeds, egocentric video, or hand-drawn sketches.
- Physical understanding: agentic multimodal hindsight labeling covering dexterity, dense value functions and progress estimation.
- Research and self-improvement: action improvement via simulation and iterative on-hardware self-improvement.
Google frames it as accelerating progress toward physical AGI.
Related event: Google Launches Gemini 4 Argon for Robotics(9 posts)→
More from Embodied
- Dyna-2 Deep Dive: World-Action Models Trained on 1M Hours of Human Video, 99.4% 24-Hour Autonomy — Latent Space · 2026-10-14
- Nature Will unveils Natura-1: 50B-parameter MoE robot foundation model trained on 1M hours of egocentric human video — rohanpaul_ai · 2026-10-14
- 15 open-source ESP32 AI gadget projects plus a step-by-step Meta Muse SDK flashing tutorial — alexcovo_eth · 2026-10-14
- Collaborative proprioception: Auki gives a phone and robot a shared sense of space — broodsugar · 2026-10-14
- Apple just released a new smart-home command center: should you buy it? — Kiteaton · 2026-10-14
- Physical Intelligence launches Pi API beta for fine-tuning π robotic foundation models — chelseabfinn · 2026-10-14