MMLU-by-task - Security Studies: leaderboard
The "Security Studies" subtask of MMLU-by-task. A single MMLU subject, reporting per-task multiple-choice accuracy from the MMLU-by-task leaderboard across a wide range of academic and professional domains.
Metric: Accuracy (%). Source: huggingface.co. Saturation forecast: Estimated already saturated. 1257 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Xwin-LM-70B-V0.1 | 81.22 | #982 |
| 2 | Llama 2 70B Base | 79.18 | #1025 |
| 3 | Llama 2 70B Chat (HF) | 78.78 | #1264 |
| 4 | Llama 2 70B Chat | 76.73 | #1237 |
| 5 | WizardLM-70B-V1.0 | 75.51 | #1236 |
| 6 | StableBeluga2 | 75.51 | #1078 |
| 7 | Mistral-7B-v0.1 | 72.65 | #1326 |
| 8 | internlm-20B | 71.43 | #1296 |
| 9 | vicuna-33B-v1.3 | 68.98 | #1347 |
| 10 | WizardLM-13B-V1.2 | 66.94 | #1425 |
| 11 | OpenHermes-13B | 64.9 | #1393 |
| 12 | trurl-2-13B-academic | 64.49 | #1413 |
| 13 | Llama 2 13B Chat Base | 64.08 | #1422 |
| 14 | falcon-40B Instruct | 63.27 | #1458 |
| 15 | trurl-2-7B | 62.45 | #1494 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=mmlu-by-task-security-studies · How It Works · Data refreshed daily, snapshot 2026-10-20.