The Bitter Lesson Comes to Robotics

Summary

Researchers evaluated nine frontier multimodal language models directly commanding physical robotic arms on six manipulation tasks from the RoboDojo benchmark, across multiple reasoning-effort and service-tier settings. The models can control real arms for a range of tasks but are not yet precise or efficient enough to finish them within a practical action budget. Performance rose between July and October 2026, with a doubling time of 53.9 days, a 4.4x increase in the best execution rate, and a 6.2x drop in the lowest execution cost.

Topics

• Robotics and embodied AI