Cyber Bench Hub (v2): leaderboard

Metric: Accuracy (%). Source: huggingface.co. 22 models tracked.

Top models

#ModelScore
1Claude 3.7 Sonnet98
2DeepSeek R198
3Phi-3 Mini 128K Instruct98
4Llama 3.3 70B Instruct97
5GPT-4o Mini97
6Phi-497
7Claude 3.5 Sonnet97
8Mistral Small 3.197
9DeepSeek V397
10Mistral Small 397
11Claude 3.5 Haiku97
12Gemini 2.0 Flash Lite97
13GPT-4.5 (Preview)97
14QwQ-32B96
15O3 Mini96

Interactive version: theaggregate.ai/benchmark?slug=cyber-bench-hub-v2 · How It Works · Data refreshed daily, snapshot 2026-09-05.