Accuracy on MMLU's high school world history questions, one of 57 subject tests of academic and professional knowledge.
unassessed
| Category | knowledge |
|---|---|
| Subcategory | Humanities |
| Page status | active |
| Metric | accuracy |
| Direction | higher_is_better |
| Unit | % |
| Dataset size | 237 |
| Dataset licence | MIT |
| Publisher | UC Berkeley |
A survey of world history across regions and eras -- civilisations, empires, revolutions, and wars -- at the level of a US high-school or AP course. 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 | 92.8 | 2024-07 |
| Meta Llama 3 70B Instruct | Meta | 92.8 | 2026-04 |
| Meta Llama 3 70B Instruct | Nous Research | 92.8 | 2026-04 |
| Mixtral 8x22B Instruct v0.1 | Mistral AI | 91.6 | 2026-04 |
| Nous Hermes 2 Yi 34B | Nous Research | 90.7 | 2024-07 |
| Yi 1.5 34B | 01.AI | 90.7 | 2026-04 |
| Yi 34B 200K | 01.AI | 90.7 | 2024-07 |
| Yi 34B Chat | 01.AI | 90.3 | 2024-07 |
| Nous Hermes 2 Mixtral 8x7B DPO | Nous Research | 89.5 | 2024-07 |
| Mixtral 8x7B v0.1 | Mistral AI | 89.0 | 2026-04 |
| Yi 1.5 34B 32K | 01.AI | 89.0 | 2024-07 |
| Yi 1.5 34B Chat | 01.AI | 88.2 | 2026-04 |
| Yi 1.5 34B Chat 16K | 01.AI | 88.2 | 2024-07 |
| Nous Hermes 2 SOLAR 10.7B | Nous Research | 87.8 | 2024-07 |
| Hermes 2 Theta Llama 3 8B | Nous Research | 86.9 | 2024-07 |
| gemma 7B it | Google DeepMind | 85.2 | 2024-07 |
| Hermes 2 Pro Llama 3 8B | Nous Research | 84.8 | 2024-07 |
| Mixtral 8x7B Instruct v0.1 | Mistral AI | 84.8 | 2026-04 |
| Yi 1.5 9B | 01.AI | 84.8 | 2024-07 |
| Yi 1.5 9B Chat 16K | 01.AI | 84.8 | 2024-07 |
| Meta Llama 3 8B Instruct | Meta | 84.4 | 2024-07 |
| Meta Llama 3 8B Instruct | Nous Research | 84.4 | 2024-07 |
| Yi 1.5 9B 32K | 01.AI | 84.4 | 2024-07 |
| Yi 1.5 9B Chat | 01.AI | 82.7 | 2024-07 |
| Meta Llama 3 8B | Meta | 82.3 | 2024-07 |
| Meta Llama 3 8B | Nous Research | 82.3 | 2024-07 |
| Yi 9B | 01.AI | 81.9 | 2024-07 |
| Phi 3 mini 128K instruct | Microsoft | 79.7 | 2024-07 |
| Phi 3 mini 4K instruct | Microsoft | 79.3 | 2024-07 |
| Yi 6B | 01.AI | 78.9 | 2024-07 |
| Yi 6B Chat | 01.AI | 78.9 | 2024-07 |
| Yi 1.5 6B Chat | 01.AI | 78.5 | 2024-07 |
| Mistral 7B v0.3 | Mistral AI | 78.1 | 2024-07 |
| mistral 7B v0.3 bnb 4bit | Unsloth | 78.1 | 2024-07 |
| Yi 1.5 6B | 01.AI | 77.2 | 2024-07 |
| Mistral 7B Instruct v0.2 | Mistral AI | 75.5 | 2024-07 |
| phi 2 | Microsoft | 74.3 | 2024-07 |
| deepseek llm 7B base | DeepSeek | 72.6 | 2024-07 |
| deepseek llm 7B chat | DeepSeek | 72.6 | 2024-07 |
| falcon 40B | TII | 71.3 | 2024-07 |
| Qwen2 1.5B Instruct | Alibaba / Qwen Team | 70.5 | 2024-07 |
| chatglm2 6B | Zhipu AI | 66.2 | 2024-07 |
| Qwen2 0.5B Instruct | Alibaba / Qwen Team | 59.1 | 2024-07 |
| gemma 2B it | Google DeepMind | 51.9 | 2024-07 |
| gemma 2B | Google DeepMind | 39.7 | 2024-07 |
| deepseek coder 6.7B instruct | DeepSeek | 37.6 | 2024-07 |
| deepseek coder 6.7B base | DeepSeek | 32.1 | 2024-07 |
| deepseek coder 1.3B base | DeepSeek | 30.8 | 2024-07 |
| deepseek coder 1.3B instruct | DeepSeek | 30.8 | 2024-07 |
| OLMo 1B hf | Allen AI | 28.3 | 2024-07 |