MMLU-by-task - High School Physics: leaderboard
The "High School Physics" 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 | Llama 2 70B Chat | 49.67 | #1237 |
| 2 | Xwin-LM-70B-V0.1 | 47.02 | #982 |
| 3 | WizardLM-70B-V1.0 | 44.37 | #1236 |
| 4 | Llama 2 70B Base | 43.05 | #1025 |
| 5 | Llama 2 70B Chat (HF) | 42.38 | #1264 |
| 6 | StableBeluga2 | 42.38 | #1078 |
| 7 | RedPajama-INCITE-Base-3B-v1 | 37.09 | #1618 |
| 8 | vicuna-33B-v1.3 | 37.09 | #1347 |
| 9 | trurl-2-13B-academic | 36.42 | #1413 |
| 10 | vicuna-13B | 35.76 | #1469 |
| 11 | OpenHermes-13B | 35.76 | #1393 |
| 12 | CodeLlama-7B-Python-hf | 35.1 | #1608 |
| 13 | trurl-2-7B | 34.44 | #1494 |
| 14 | falcon-40B Instruct | 34.44 | #1458 |
| 15 | stablelm-base-alpha-7B | 34.44 | #1629 |
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-high-school-physics · How It Works · Data refreshed daily, snapshot 2026-10-20.