Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
Mathematical Problem Solving on MATH-Perturb MP-hard
Loading...
56.3
Exact Match Accuracy
DeepMath (RULES)
6.692
19.571
32.45
45.329
Feb 1, 2026
Exact Match Accuracy
Updated 4d ago
Evaluation Results
Method
Method
Links
Exact Match Accuracy
DeepMath (RULES)
Optimization=RULES
2026.02
56.3
DeepMath (GRPO)
Optimization=GRPO
2026.02
54.5
DeepMath (vanilla)
Optimization=vanilla
2026.02
53
Qwen2.5 (RULES)
Optimization=RULES
2026.02
35.1
Qwen2.5 (vanilla)
Optimization=vanilla
2026.02
33.7
Qwen2.5 (GRPO)
Optimization=GRPO
2026.02
33.7
Llama-3.1 (vanilla)
Optimization=vanilla
2026.02
13.3
Llama-3.1 (GRPO)
Optimization=GRPO
2026.02
13.3
Llama-3.1 (RULES)
Optimization=RULES
2026.02
12.9
OLMO2 (RULES)
Optimization=RULES
2026.02
10.4
OLMO2 (GRPO)
Optimization=GRPO
2026.02
9.3
OLMO2 (vanilla)
Optimization=vanilla
2026.02
8.6
Feedback
Search any
task
Search any
task