Gemini 4 Argon taught a humanoid robot cornhole from one video, landing 7 straight shots
141_1337 · reddit · 2026-10-14
A Reddit post showcases Google's latest robotics demo: the humanoid robot Apollo learned to play cornhole from a single human video using Gemini 4 Argon.
The pipeline has three steps:
- The model built its own physics simulator
- It trained the throwing policy in simulation
- The policy was refined on the real robot
Apollo then landed 7 shots in a row on the real board. According to Google, this video-to-sim-to-real workflow compresses weeks of robotics engineering into hours.
More from Embodied
- 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
- Gemini 4 Argon goes GA, reportedly controls full-body humanoid with dexterous hands — evijit · 2026-10-14
- VPP2: video-prediction robot policy beats Cosmos3-64B by 11%, tops RoboDojo — lucy_x_shi · 2026-10-14
- China reportedly shipped 97% of the world's humanoid robots in H1 2026 — Polymarket · 2026-10-14
- China's Tiangong Ultra robot beat Bolt's 100m record — and exposes America's robotics supply-chain dilemma — mattsheehan88 · 2026-10-14