external benchmark
CyScenarioBench solved leaderboard
CyScenarioBench Scenarios Ever Solved. Every model the catalog carries a published CyScenarioBench solved value for, ranked by that value.
Categoryexternal
MeasureScenarios solved at least once
Tasks11 cyber scenarios
DifficultyLong-horizon cybersecurity
Published byAssessing GPT-5.6 Sol
Number of CyScenarioBench scenarios completed successfully in at least one run.
CyScenarioBench solved ranking
2 models with a published CyScenarioBench solved value, ordered by that value, highest first. Models the source has not scored on this benchmark are not listed — they are unmeasured, not last.
| Rank | Model | Provider | Scenarios solved at least once |
|---|---|---|---|
| 1 |  GPT-6 Astra | OpenAI | 9 |
| 2 | GPT-5.6 Sol | OpenAI | 7 |
Evidence key: Observed
Rows are ordered by the value Assessing GPT-5.6 Sol published, highest first. The source does not state whether a higher value is the better result, so this page does not either: for a benchmark that measures a rate of failure, read the table from the bottom.