HIDBench - NodLink HW20: leaderboard
HIDBench (2026) tests LLMs as host-based intrusion detectors: the model reads windows of serialized provenance-graph events from system logs and must flag the malicious ones, here on the NodLink simulated HW20 host logs. Matthews correlation coefficient of malicious-event detection times 100 (-100 to 100); higher is better.
Metric: Matthews correlation coefficient of malicious-event detection times 100 (-100 to 100), from the HIDBench pipeline (attack-centric windows of host provenance-graph events serialized for the LLM, high-confidence evidence identification, k-hop context expansion, then attack narrative and indicator-of-compromise output) on the NodLink simulated-data HW20 host logs (1,217 events, 43 malicious); higher is better. Source: arxiv.org. Saturation forecast: Around February 2027. 9 models tracked. Updated 7 Oct 2026.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Qwen 3.6 Plus | 73.9 | #245 |
| 2 | DeepSeek V3.2 | 69.7 | #425 |
| 3 | Claude Sonnet 4.5 | 69.5 | #267 |
| 4 | Claude Opus 4.6 | 63.3 | #114 |
| 5 | Claude Sonnet 4 | 55.7 | #414 |
| 6 | GPT-5.2 | 52.9 | #190 |
| 7 | GPT-4.1 | 48.7 | #506 |
| 8 | GPT-OSS-120B | 46.3 | #712 |
| 9 | Gemini 2.5 Flash | 32.8 | #466 |
No result here: #2 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.
Interactive version: theaggregate.ai/benchmark?slug=hidbench-nodlink-hw20 · How It Works · Data refreshed daily, snapshot 2026-11-22.