Are Frontier VLM Agents Ready to Be Robot Generalists? An Empirical Study with the Embodied Agent Arena

A study introducing the Embodied Agent Arena evaluated ten vision-language model agents across five embodied task categories — Geometry, Spatial Reasoning, Affordance, Task Planning, and Manipulation — to test whether frontier VLMs can serve as robot generalists. The research asks how far these models meet the demands of interpreting a scene, acting on it, and satisfying every condition of a task.

A robot generalist must interpret a scene, act on it, and satisfy every condition of a task. How far do frontier vision-language models VLMs meet these demands? We introduce Embodied Agent Arena, evaluating ten VLM agents across Geometry, Spatial Reasoning, Affordance, Task Planning, and Manipulat