Benchmark Radar
AI BENCHMARK PROFILE

WorldExam

General AIMultimodal Perception

WorldExam evaluates controllable video generation models as world models across four levels: Visual Quality, Control Adherence, Spatial Consistency, and World Reactivity, with 1,474 cases across eight tasks.

Released
2026-08-03
Readiness
Runnable
Primary field
General AI

Why it matters

Fills the gap in evaluating inherent world reactivity beyond visual quality and instruction fulfillment, providing a hierarchical diagnostic for model comparison.

Motivation

Controllable video generation models are increasingly being developed as world models.

Primary resources

Benchmark Radar records only publicly supported details and links back to primary sources for verification.