Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

R1-Onevision-Bench

Benchmarks

Task NameDataset NameSOTA ResultTrend
Multimodal ReasoningR1-Onevision-Bench (Overall)
Accuracy39.2
23
Multimodal ReasoningR1-Onevision-Bench Physics
Accuracy34.9
8
Multimodal ReasoningR1-Onevision-Bench Math
Accuracy25.7
8
Multimodal ReasoningR1-Onevision-Bench Deduction
Accuracy27.8
8
Showing 4 of 4 rows