HIDBench: leaderboard
Recent benchmark efforts have advanced the evaluation of large language models (LLMs) in cybersecurity, including tasks such as penetration testing and vulnerability identification. However, a critical cybersecurity task, namely intrusion detection from system logs, remains unexplored. In this work, we present a new benchmark to assess LLMs' capabilities in supporting host-based intrusion detectio
Metric: DARPA E3 CADETS MCC (self-reported). Source: benchmarklist.com. Saturation forecast: Around August 2027. 9 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Opus 4.6 | 60.2 | #63 |
| 2 | Claude Sonnet 4 | 56.6 | #199 |
| 3 | Claude Sonnet 4.5 | 56.1 | #142 |
| 4 | GPT-4.1 | 53.4 | #248 |
| 5 | GPT-5.2 | 51 | #110 |
| 6 | GPT-OSS-120B | 42.9 | #341 |
| 7 | DeepSeek V3.2 | 42.8 | #205 |
| 8 | Qwen 3.6 Plus | 36.4 | #120 |
| 9 | Gemini 2.5 Flash | 27.9 | #245 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=hidbench · How It Works · Data refreshed daily, snapshot 2026-10-20.