ExploitGym (v1; selected subset): leaderboard
Successful exploits using the intended vulnerability on a selected subset of ExploitGym v1 with the DoGNAVY agent. Scores are counts, not percentages or coverage of the full 869-task v1 suite. The source does not specify a timeout for this row; the time budget of another run is not assumed.
Metric: Successful intended exploits (#); v1; selected subset. Source: www.cybergym.io. 1 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GLM-5.2 | 79 | #78 |
Interactive version: theaggregate.ai/benchmark?slug=exploitgym-v1-selected-subset · How It Works · Data refreshed daily, snapshot 2026-10-28.