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

Process-outcome alignment on PRIME Math

92.44Outcome Score

Claude-Sonnet-4.5-thinking

64.484871.74247986.2576Feb 12, 2026
Updated 4d ago

Evaluation Results

MethodLinks
2026.02
92.4481
2026.02
92.1186.11
2026.02
92.1186
2026.02
91.7882.56
2026.02
91.5980.89
2026.02
91.3385.78
2026.02
91.2278.33
2026.02
91.1186.67
2026.02
90.8975.67
2026.02
90.7881.89
2026.02
90.7881
2026.02
90.6783.89
2026.02
90.5681.67
2026.02
90.4479.67
2026.02
90.2281
2026.02
90.2274
2026.02
9074.78
2026.02
89.6775.56
2026.02
89.4474.78
2026.02
89.3375.33
2026.02
89.2276.22
2026.02
88.8979.33
2026.02
88.3367.33
2026.02
88.1171.11
2026.02
87.7872.89
2026.02
87.5672.44
2026.02
87.5668.78
2026.02
87.1167.44
2026.02
86.3369.56
2026.02
86.2266.33
2026.02
81.4460
2026.02
81.2264.78
2026.02
76.7857.78
2026.02
65.5663.67