List your company

ResearchFormerly WizWand

Open-ended evaluation

Benchmarks

DatasetSOTA methodMetricTrendResultsLast updated

8.93p+ Score

39

Apr 15, 2026

RadBench open-ended reduced split n=69

2RadGraph F1

8

Jul 8, 2026

FineInstructions

3.9Likert Score

8

Feb 26, 2026

DomainBench (test)

27.98Geography Score

7

Apr 16, 2026