LLM Stats (ExploitGym): leaderboard
LLM Stats mirror of ExploitGym, 898 exploit-development tasks built from real vulnerabilities in C/C++ userspace projects, Google’s V8 engine and the Linux kernel. The agent works a containerised target over multiple rounds under configurable mitigations, and a run only counts when it captures the flag and a judge confirms the exploit used the intended vulnerability rather than some other way in. UC Berkeley, MPI-SP, UC Santa Barbara and Arizona State, 2026.
Metric: Score (%). Source: llm-stats.com. Saturation forecast: Around December 2027. 8 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-6 Astra | 42.4 | #6 |
| 2 | GPT-5.6 Sol | 33.7 | #19 |
| 3 | GPT-5.6 Terra | 23.2 | #41 |
| 4 | MiMo-V2.6-Pro | 17.8 | #42 |
| 5 | DeepSeek V4.1 Flash | 15.3 | #36 |
| 6 | GLM-5.3 | 15 | #39 |
| 7 | GPT-5.6 Luna | 12.4 | #68 |
| 8 | MiMo-V2.6-Flash | 6 | #70 |
No result here: #3 Claude Opus 5.5, #7 GPT-6.1 Sol, #8 Gemini 4 Argon.
Interactive version: theaggregate.ai/benchmark?slug=llm-stats-exploitgym · How It Works · Data refreshed daily, snapshot 2026-10-20.