Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

SynthWorlds

Benchmarks

Task NameDataset NameSOTA ResultTrend
ReasoningSynthWorlds 2026 (test)
Score0.48
15
ReasoningSynthWorlds
Reasoning Score (SynthWorlds)48
15
Multi-hop Question AnsweringSynthWorlds-RM
Exact Match (EM)26.9
10
Showing 3 of 3 rows