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

Derivation on OneReason-Bench

28.6I2I Score

OneReason RFT thinking

2.42329.219116.01522.8109Jun 4, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.06
28.6
2026.06
28.36
2026.06
25.54
2026.06
23.99
2026.06
23.88
2026.06
16.21
2026.06
3.43