TACTL — leaderboard
Offensive cybersecurity knowledge test evaluating LLMs on tactical cyber operations, threat intelligence, and adversarial techniques.
Metric: Accuracy (%). Source: huggingface.co. Status: saturated. 8 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | DeepSeek R1 | 95.9 |
| 2 | DeepSeek V3 | 93.15 |
| 3 | Llama 3.1 405B | 90.9 |
| 4 | GPT-4o | 89.25 |
| 5 | Qwen 2.5 72B Instruct | 88.75 |
| 6 | Llama 3.3 70B | 79.35 |
| 7 | Mixtral 8x22B | 62.5 |
Interactive version: theaggregate.ai/benchmark?slug=tactl · How the rankings work · Data refreshed daily, snapshot 2026-07-22.