Accuracy on MMLU's high school statistics questions, one of 57 subject tests of academic and professional knowledge.
unassessed
| Category | knowledge |
|---|---|
| Subcategory | STEM |
| Page status | active |
| Metric | accuracy |
| Direction | higher_is_better |
| Unit | % |
| Dataset size | 216 |
| Dataset licence | MIT |
| Publisher | UC Berkeley |
Introductory descriptive and inferential statistics at AP level: probability, distributions, hypothesis testing, confidence intervals, and basic regression. Framed as four-option multiple-choice questions and scored zero-shot or few-shot by exact match against the labelled option, as one of the 57 subject subsets that make up the MMLU benchmark.
Four-option multiple-choice question answering (A-D), one correct answer, zero-shot or few-shot.
| Model | Provider | Score | Card as of |
|---|---|---|---|
| Meta Llama 3 70B | Meta | 73.6 | 2024-07 |
| Meta Llama 3 70B Instruct | Meta | 73.6 | 2026-04 |
| Meta Llama 3 70B Instruct | Nous Research | 73.6 | 2026-04 |
| Yi 1.5 34B | 01.AI | 72.7 | 2026-04 |
| Yi 1.5 34B 32K | 01.AI | 71.3 | 2024-07 |
| Mixtral 8x22B Instruct v0.1 | Mistral AI | 69.4 | 2026-04 |
| Yi 1.5 9B Chat | 01.AI | 68.5 | 2024-07 |
| Yi 9B | 01.AI | 68.5 | 2024-07 |
| Yi 1.5 34B Chat | 01.AI | 68.1 | 2026-04 |
| Yi 1.5 34B Chat 16K | 01.AI | 68.1 | 2024-07 |
| Yi 34B 200K | 01.AI | 67.1 | 2024-07 |
| Nous Hermes 2 Mixtral 8x7B DPO | Nous Research | 66.2 | 2024-07 |
| Nous Hermes 2 Yi 34B | Nous Research | 66.2 | 2024-07 |
| Yi 1.5 9B 32K | 01.AI | 66.2 | 2024-07 |
| Yi 1.5 9B | 01.AI | 65.3 | 2024-07 |
| Yi 1.5 9B Chat 16K | 01.AI | 65.3 | 2024-07 |
| Mixtral 8x7B v0.1 | Mistral AI | 64.8 | 2026-04 |
| Phi 3 mini 128K instruct | Microsoft | 64.8 | 2024-07 |
| Yi 34B Chat | 01.AI | 63.4 | 2024-07 |
| Phi 3 mini 4K instruct | Microsoft | 61.6 | 2024-07 |
| Mixtral 8x7B Instruct v0.1 | Mistral AI | 59.7 | 2026-04 |
| gemma 7B it | Google DeepMind | 57.9 | 2024-07 |
| Meta Llama 3 8B | Meta | 57.9 | 2024-07 |
| Meta Llama 3 8B | Nous Research | 57.9 | 2024-07 |
| Yi 1.5 6B | 01.AI | 57.4 | 2024-07 |
| Meta Llama 3 8B Instruct | Meta | 53.7 | 2024-07 |
| Meta Llama 3 8B Instruct | Nous Research | 53.7 | 2024-07 |
| Yi 1.5 6B Chat | 01.AI | 53.2 | 2024-07 |
| Nous Hermes 2 SOLAR 10.7B | Nous Research | 52.8 | 2024-07 |
| Mistral 7B v0.3 | Mistral AI | 51.9 | 2024-07 |
| mistral 7B v0.3 bnb 4bit | Unsloth | 51.9 | 2024-07 |
| Yi 6B | 01.AI | 51.9 | 2024-07 |
| Yi 6B Chat | 01.AI | 51.9 | 2024-07 |
| Hermes 2 Pro Llama 3 8B | Nous Research | 50.9 | 2024-07 |
| Hermes 2 Theta Llama 3 8B | Nous Research | 50.9 | 2024-07 |
| falcon 40B | TII | 47.7 | 2024-07 |
| phi 2 | Microsoft | 47.2 | 2024-07 |
| Mistral 7B Instruct v0.2 | Mistral AI | 44.4 | 2024-07 |
| Qwen2 1.5B Instruct | Alibaba / Qwen Team | 44.4 | 2024-07 |
| deepseek llm 7B base | DeepSeek | 43.5 | 2024-07 |
| deepseek llm 7B chat | DeepSeek | 43.5 | 2024-07 |
| Qwen2 0.5B Instruct | Alibaba / Qwen Team | 38.9 | 2024-07 |
| deepseek coder 6.7B base | DeepSeek | 37.0 | 2024-07 |
| deepseek coder 6.7B instruct | DeepSeek | 36.6 | 2024-07 |
| OLMo 1B hf | Allen AI | 36.6 | 2024-07 |
| gemma 2B | Google DeepMind | 35.2 | 2024-07 |
| chatglm2 6B | Zhipu AI | 33.8 | 2024-07 |
| deepseek coder 1.3B base | DeepSeek | 25.5 | 2024-07 |
| deepseek coder 1.3B instruct | DeepSeek | 25.5 | 2024-07 |
| gemma 2B it | Google DeepMind | 20.4 | 2024-07 |