CyberGym — leaderboard
Cybersecurity agent benchmark for discovering, exploiting, and reasoning about vulnerabilities in controlled challenge environments.
Metric: Score (self-reported). Source: benchmarklist.com. Status: saturation imminent. 8 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude Mythos 5 | 83.8 |
| 2 | Claude Mythos Preview | 83.1 |
| 3 | GPT-5.5 | 81.8 |
| 4 | GPT-5.4 | 79 |
| 5 | Claude Opus 4.8 | 78.1 |
| 6 | Claude Opus 4.6 | 73.8 |
| 7 | Claude Opus 4.7 | 73.1 |
| 8 | Claude Sonnet 4.6 | 65.2 |
Interactive version: theaggregate.ai/benchmark?slug=cybergym · How the rankings work · Data refreshed daily, snapshot 2026-07-22.