CyberCertBench - Fortinet ICS/SCADA: leaderboard
CyberCertBench (2026) certification knowledge test of 43 Fortinet ICS/SCADA questions (conceptual operational-technology security), asked as 5-shot multiple choice at temperature 0. Exact accuracy (%): a question counts only when all correct options and no wrong one are chosen; higher is better.
Metric: Accuracy (%) on 43 Fortinet ICS/SCADA questions (conceptual operational-technology security): 5-shot multiple-choice question answering at temperature 0, scored by exact accuracy (correct only when the model selects all correct options and no incorrect one), mean accuracy (%) as printed in the appendix; questions scraped from public certification question banks and checked against authoritative documentation; higher is better. Source: arxiv.org. Saturation forecast: Estimated already saturated. 45 models tracked. Updated 3 Nov 2026.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5 | 88.4 | #185 |
| 2 | GPT-OSS-120B | 88.4 | #712 |
| 3 | Mistral Small 3.1 | 88.4 | #1734 |
| 4 | Qwen 2.5 72B | 87.4 | #1189 |
| 5 | GPT-4o | 86.5 | #754 |
| 6 | GPT-4.1 | 86 | #506 |
| 7 | Mistral Large 2 (Nov) Instruct (2411) | 86 | #1224 |
| 8 | Gemini 2.5 Pro (Preview 05-06) | 86 | #278 |
| 9 | Qwen 2 72B | 86 | #1028 |
| 10 | Claude Haiku 4.5 | 83.7 | #587 |
| 11 | GPT-5 Mini | 83.7 | #373 |
| 12 | Kimi K2 | 83.7 | #448 |
| 13 | Claude 3.5 Sonnet | 83.7 | #755 |
| 14 | Qwen 3 32B | 81.4 | #983 |
| 15 | Gemini 2.5 Flash Lite | 81.4 | #972 |
No result here: #2 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.
Interactive version: theaggregate.ai/benchmark?slug=cybercertbench-fortinet-ics-scada · How It Works · Data refreshed daily, snapshot 2026-11-22.