CyberGym-E2E (615 tasks) - PoC Generation: leaderboard

Metric: End-to-end stage S1 rate (%) on the initial 615-task release (OSS-Fuzz vulnerabilities in C/C++ projects): the agent-written PoC input crashes the unpatched build under a sanitizer, with no ground-truth vulnerability data given; with a $10 cost cap and a 90-minute limit per task; higher is better. Source: arxiv.org. Saturation forecast: Around December 2027. 5 models tracked.

Top models

#ModelScore
1GPT-5.2 Codex30.2
2Gemini 3 Pro29.6
3Claude Opus 4.524.9
4Claude Sonnet 4.518.1

Interactive version: theaggregate.ai/benchmark?slug=cybergym-e2e-615-tasks-poc-generation · How It Works · Data refreshed daily, snapshot 2026-09-29.