Benchmark suiteResults available
WBench
Multi-turn interactive world-model evaluation across quality, setting, interaction, consistency and physical behavior.
60 published resultsExplore benchmark
Rather
The benchmark catalog
Find the right test.
Start with the capability you care about. Explore what each benchmark measures, which models it covers, and where the results come from.
Control & interaction
How generated worlds respond to actions, navigation, and user control.
1 suite · 60 published results
Categories group related tests. Each suite keeps its own metrics and protocols; category counts include only suites with published results.
Coverage in progress
Pending integrations
Listed for visibility. These suites do not yet have published results on Rather.
Benchmark suiteIntegration pending
iWorld-Bench
Evaluation of camera control, visual quality and memory in interactive world models.
No published results yetView integration