List your company

ResearchFormerly WizWand

Hierarchical Task Decomposition

Benchmarks

DatasetSOTA methodMetricTrendResultsLast updated

MagicEval-Plan Context Inheritance 3

97.6Step Score

14

Feb 26, 2026

MagicEval-Plan Condition 3

97.5Step Count

14

Feb 26, 2026

MagicEval-Plan Dependency 3

Qwen3-30B-A3B-Instruct

16Step Count

14

Feb 26, 2026

MagicEval-Plan General 3

Qwen3-30B-A3B-Instruct

35.1Step Count

14

Feb 26, 2026