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

Math Reasoning on MultiArith (Accuracy and Reasoning Length)

99.3Accuracy

CoT-SFT

14.22836.31458.480.486Jan 21, 2026Feb 16, 2026Mar 15, 2026Apr 10, 2026May 7, 2026Jun 2, 2026Jun 29, 2026
Updated 25d ago

Evaluation Results

MethodLinks
2026.06
99.369.4
2026.01
98.359.1
2026.01
97.232
2026.01
9568
2026.06
9568
2026.06
94.44.2
2026.06
93.74.1
2026.01
89.22.28
2026.06
89.22.3
2026.01
873.23
2026.06
873.2
2026.01
86.182.3
2026.01
85.60
2026.06
79.43.4
2026.01
68.332
2026.01
62.232
2026.06
62.232
2026.06
52.96
2026.01
52.80
2026.01
41.70
2026.06
41.70
2026.01
41.46
2026.06
41.46
2026.01
38.20
2026.06
38.20
2026.06
32.20
2026.01
19.26
2026.06
19.26
2026.06
17.56