Skip to content

Benchmark

Hy-Math

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

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

Score distribution

60.9%
Mean
60.9%
Median
0%
Std dev
60.9%
10th percentile
60.9%
90th percentile
60.9%
Min

Leaderboard

#ModelScore
1Hy3 (Deep Infra)Hy3 official Benchmark Appendix · highest reasoning tier unless noted · internal· Reported by Tencent Hunyuan· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.60.9%

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.