the ai forger
Models
Organizations
Providers
Benchmarks
Compare

MATH-500

math
text
About

MATH-500 benchmark

Evaluation Stats
Total Models22
Organizations8
Verified Results0
Self-Reported22
Benchmark Details
Max Score1
Language
en
Performance Overview
Score distribution and top performers

Score Distribution

22 models
Top Score
97.4%
Average Score
91.3%
High Performers (80%+)
20

Top Organizations

#1Moonshot AI
2 models
96.8%
#2NVIDIA
3 models
96.3%
#3Anthropic
1 model
96.2%
#4Microsoft
1 model
94.6%
#5DeepSeek
10 models
92.6%
Leaderboard
Top 20 models ranked by performance
1
97.4%
Raw: 0.974
Self-reported
2
97.3%
Raw: 0.973
Self-reported
3
97.0%
Raw: 0.97
Self-reported
4
96.6%
Raw: 0.966
Self-reported
5
96.2%
Raw: 0.962
Self-reported
6
96.2%
Raw: 0.962
Self-reported
7
95.9%
Raw: 0.959
Self-reported
8
95.4%
Raw: 0.954
Self-reported
9
94.6%
Raw: 0.946
Self-reported
10
94.5%
Raw: 0.945
Self-reported
11
94.3%
Raw: 0.943
Self-reported
12
94.0%
Raw: 0.94
Self-reported
13
93.9%
Raw: 0.939
Self-reported
14
92.8%
Raw: 0.928
Self-reported
15
90.6%
Raw: 0.906
Self-reported
16
90.6%
Raw: 0.906
Self-reported
17
90.2%
Raw: 0.902
Self-reported
18
90.0%
Raw: 0.9
Self-reported
19
89.1%
Raw: 0.891
Self-reported
20
83.9%
Raw: 0.839
Self-reported
Showing top 20 of 22 models