CyberCertBench - Cisco CCNx: leaderboard
Metric: Accuracy (%) on 200 Cisco networking certification questions (IT networking and security principles): 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.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 2.5 Pro (Preview 05-06) | 91.5 |
| 2 | Claude Sonnet 4.5 | 89.5 |
| 3 | Kimi K2 | 86.5 |
| 4 | Claude 3.5 Sonnet | 86.5 |
| 5 | GPT-4o | 85.8 |
| 6 | GPT-5 | 85.5 |
| 7 | Mistral Medium 3.1 | 84.5 |
| 8 | GPT-4.1 | 84 |
| 9 | Claude Haiku 4.5 | 82 |
| 10 | DeepSeek V3 (0324) | 82 |
| 11 | GPT-5 Mini | 81 |
| 12 | Gemini 2.0 Flash | 80.8 |
| 13 | Llama 3.1 405B | 80.6 |
| 14 | Qwen 3 235B A22B | 80.5 |
| 15 | Mistral Small 3.1 | 79.5 |
Interactive version: theaggregate.ai/benchmark?slug=cybercertbench-cisco-ccnx · How It Works · Data refreshed daily, snapshot 2026-10-07.