Skip to content

Benchmark

MathVision · With CI

Math benchmarks measure how well models solve maths problems. Scores are percentages. Higher is better.

Models measured
8
Scores recorded
9
Sources
1
IQ weight
No weight published

Score distribution

94.9%
Mean
95.7%
Median
2.7%
Std dev
93.2%
10th percentile
97.7%
90th percentile
88.4%
Min
88.4%97.8%
The band covers the middle 80% of scores. The line marks the median.

Leaderboard

#ModelScore
1Kimi K3Python tool · three-run average · max effort· Reported by Moonshot AI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.97.8%
2Qwen3.8 MaxQwen3.8-Max release vision/video table · with Code Interpreter· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.97.7%
3Qwen3.8 Omni FlashWith Code Interpreter· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.96.2%
4Qwen 3.8 Flash NextQwen3.8-Flash-Next release/model card · with Code Interpreter· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.95.7%
5Qwen3.8 FlashWith Code Interpreter· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.95.7%
6Qwen 3.8 27BQwen3.8-27B model card · with Code Interpreter· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.94.6%
7Qwen 3.8 27BWith Code Interpreter· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.94.6%
8Kimi K2.6Kimi K2.6 official benchmark table · thinking · temp 1.0 · top_p 1.0 · 262,144 context · with Python · max 50 steps · avg@3· Reported by Moonshot AI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.93.2%
9Qwen3.7 PlusWith Code Interpreter· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.88.4%

Data credits: BenchBoard by SYuan03 (github.com/SYuan03/benchboard). MIT covers its code and data organization; no results licence is inferred. Numeric relays retain the original stating-source links.