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

Perception on OneReason-Bench

36.91Item Understanding

OneReason SFT thinking

22.422826.183929.94533.7061Jun 4, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.06
36.911.0664.6
2026.06
36.843.9566.55
2026.06
36.825.2467.25
2026.06
36.781.3565.65
2026.06
35.415.2163.9
2026.06
26.48056.45
2026.06
22.9800.4