MMLU-by-task - High School Macroeconomics — leaderboard
Metric: Accuracy (%). Source: huggingface.co. 1257 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Llama 2 70B Base | 74.1 |
| 2 | StableBeluga2 | 70.77 |
| 3 | falcon-180B | 70.77 |
| 4 | Llama 2 70B Chat | 70.26 |
| 5 | Mistral-7B-v0.1 | 66.67 |
| 6 | LLaMA-65B | 66.15 |
| 7 | Llama 2 70B Chat (HF) | 64.1 |
| 8 | Llama 2 70B Chat GPTQ | 62.82 |
| 9 | vicuna-33B-v1.3 | 58.21 |
| 10 | LLaMA-30B | 57.18 |
| 11 | internlm-20B Chat | 56.92 |
| 12 | falcon-40B Instruct | 55.13 |
| 13 | vicuna-13B-v1.5 | 54.62 |
| 14 | OpenHermes-13B | 54.1 |
| 15 | trurl-2-13B-academic | 53.08 |
Interactive version: theaggregate.ai/benchmark?slug=mmlu-by-task-high-school-macroeconomics · How the rankings work · Data refreshed daily, snapshot 2026-07-22.