Skip to content

Benchmark

SkillsBench · Text-only (79 tasks)

Scores are percentages. Higher is better.

Models measured
8
Scores recorded
8
Sources
1
IQ weight
0.00

Score distribution

61.4%
Mean
63.3%
Median
4.8%
Std dev
55.3%
10th percentile
65%
90th percentile
51.9%
Min
51.9%66.7%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 0.00 for chat models, measured over 5 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
1Qwen3.8 MaxHy4 official Benchmark Appendix · highest available reasoning setting · Claude Code · 79 self-contained text-only tasks · three-run average· Reported by Tencent Hy· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.66.7%
2DeepSeek V4 Pro 0813Hy4 official Benchmark Appendix · highest available reasoning setting · Claude Code · 79 self-contained text-only tasks · three-run average· Reported by Tencent Hy· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.65%
3Claude Opus 5Hy4 official Benchmark Appendix · highest available reasoning setting · Claude Code · 79 self-contained text-only tasks · three-run average· Reported by Tencent Hy· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.63.7%
4GLM 5.3Hy4 official Benchmark Appendix · highest available reasoning setting · Claude Code · 79 self-contained text-only tasks · three-run average· Reported by Tencent Hy· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.63.3%
5Hy4 previewHy4 official Benchmark Appendix · highest available reasoning setting · Claude Code · 79 self-contained text-only tasks · three-run average· Reported by Tencent Hy· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.62.9%
6GPT-5.6 SolHy4 official Benchmark Appendix · highest available reasoning setting · Claude Code · 79 self-contained text-only tasks · three-run average· Reported by Tencent Hy· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.62.5%
7Hy3 (Deep Infra)Hy3 official Benchmark Appendix · highest reasoning tier unless noted · Claude Code · 79 text-only self-contained tasks · three-run average· Reported by Tencent Hunyuan, Tencent Hy· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.55.3%
8Kimi K3Hy4 official Benchmark Appendix · highest available reasoning setting · Claude Code · 79 self-contained text-only tasks · three-run average· Reported by Tencent Hy· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.51.9%

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.