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

Mathematical Reasoning on MATH-500 (AUCOAA)

91.9AUCOAA

Normalized-Length

Updated 4d ago

Evaluation Results

Method	Links
Normalized-Length 2026.01		91.9
Format-Adaptive-Answer 2026.01		91.2
Adaptive-Answer 2026.01		90.3
TWYN 2026.01		89.6
Hard-Length 8k → 4k 2026.01		89.1
SFT 2026.01		87.9
Hard-Length 8k 2026.01		87.9
Soft-Length 2026.01		87.9
Hard-Length 16k 2026.01		83.4
Base model 2026.01		81.7
No-Thinking 2026.01		81.3