Meta's FAIR Releases InteractiveWorld, a Simulator That Scores AI Assistants on World State

Meta's FAIR releases InteractiveWorld, a simulator where an AI assistant sees a human's first-person view and issues short commands in 44 BEHAVIOR-1K scenes. It ships InteractiveBench, 120 frozen multi-turn tasks across six categories, scoring goals against world state rather than text. The code is under CC BY-NC 4.0, and the human body and motion data require separate licenses.

• The README says InteractiveWorld needs an NVIDIA A100 80 GB GPU, about 4 GB per simulator worker, and roughly 100 GB of disk.
• The asset preparation step bakes 6,831 BEHAVIOR objects and finishes in about two minutes with eight workers, per the README.
• The simulator supports 36 human action families, including [Grab], [PutInside], [OpenDoor], [Sit] and [AnswerPhone].