Vibe Robotics: Artifact Arena makes frontier models engineer competing robots

kaixhin · x · 2026-10-06

• Researchers introduce "Vibe Robotics": agents must produce fully functioning embodied robots (hardware + software) that solve tasks in grounded physics environments, launching the Artifact Arena benchmark.
• Frontier models design robot bodies, controllers, and competitive strategies; creations fight in a Last Bot Standing ring game (push-out or pin), with outcomes verified by a physics simulator.
• It measures hardware–software co-design, 3D physical reasoning/world modeling, 3D design & coding, and gameplay strategy, via three harnesses: sampling, verifier-grounded refinement, and design lab.
• The leaderboard accepts community submissions and stays non-saturating: each new design competes against the best artifacts built so far. Paper and site available.

Original post →

More from Embodied

• Stanford Open-Sources DITTO-X: Force-Feedback Teleop With Reverse Human Intervention — CyberRobooo · 2026-10-06
• $3,500 local AI PC under fire: only 24GB VRAM and priced below its own parts cost — BLUECOW009 · 2026-10-06
• Actual raw lidar image from Waymo's 6th-gen sensor suite shared online — reed · 2026-10-06
• RT-SAFE benchmark: 94.1% of 8 frontier VLMs reach goals, only 0.7% finish with zero safety events — Lianhuiq · 2026-10-06
• PointWAM: 3D world action model beats dexterous manipulation SOTA by 11.7 points — Chunghyun Park · 2026-10-06
• JEPA-TTT cuts latent world-model prediction error 83% under dynamics shifts — JHU · 2026-10-06

Embodied channel →