Skip to content

Benchmark

Capture-the-Flag Challenges

Scores are percentages. Higher is better.

Models measured
5
Scores recorded
6
Sources
1
IQ weight
No weight published

Score distribution

87.9%
Mean
88.1%
Median
5.9%
Std dev
85.2%
10th percentile
96.7%
90th percentile
77.6%
Min
77.6%96.7%
The band covers the middle 80% of scores. The line marks the median.

Leaderboard

#ModelScore
1GPT-5.6 SolOpenAI GPT-5.6 launch table· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.96.7%
2GPT-5.6 TerraOpenAI GPT-5.6 launch table· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.91.8%
3GPT 5.5expanded internal hardest CTF set · xhigh· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.88.1%
4GPT 5.5OpenAI GPT-5.6 launch table· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.88.1%
5GPT-5.6 LunaOpenAI GPT-5.6 launch table· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.85.2%
6GPT-5.3 CodexCybersecurity CTF challenges · xhigh· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.77.6%

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.