MMLU: Nutrition

Accuracy on MMLU's nutrition questions, one of 57 subject tests of academic and professional knowledge.

Also known as: nutrition

unassessed

This page is a discovery lead. Nobody has yet assessed it against the catalogue contract, so it carries no disposition. Absence of evidence here is not evidence of staleness.
Categoryknowledge
SubcategoryOther
Page statusactive
Metricaccuracy
Directionhigher_is_better
Unit%
Dataset size306
Dataset licenceMIT
PublisherUC Berkeley

What it measures

Dietary science: macronutrients and micronutrients, metabolism, diet-related disease, and nutrition guidelines, at an introductory course level. 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.

Task format

Four-option multiple-choice question answering (A-D), one correct answer, zero-shot or few-shot.

Models reporting this benchmark

These figures come from the model cards, which carry one collection date per card and no per-score attribution. They are shown as reported, not as verified evidence.
ModelProviderScoreCard as of
Mixtral 8x22B Instruct v0.1Mistral AI87.92026-04
Meta Llama 3 70BMeta87.62024-07
Meta Llama 3 70B InstructMeta87.62026-04
Meta Llama 3 70B InstructNous Research87.62026-04
Yi 34B 200K01.AI84.62024-07
Nous Hermes 2 Yi 34BNous Research84.32024-07
Yi 1.5 34B01.AI84.32026-04
Yi 1.5 34B 32K01.AI83.72024-07
Yi 34B Chat01.AI83.02024-07
Yi 1.5 34B Chat01.AI82.72026-04
Yi 1.5 34B Chat 16K01.AI82.72024-07
Mixtral 8x7B v0.1Mistral AI82.42026-04
Mixtral 8x7B Instruct v0.1Mistral AI82.02026-04
Nous Hermes 2 Mixtral 8x7B DPONous Research80.72024-07
Yi 1.5 9B01.AI78.42024-07
Yi 1.5 9B 32K01.AI78.12024-07
Nous Hermes 2 SOLAR 10.7BNous Research77.82024-07
Phi 3 mini 128K instructMicrosoft77.82024-07
Yi 1.5 9B Chat 16K01.AI77.52024-07
gemma 7B itGoogle DeepMind76.82024-07
Meta Llama 3 8BMeta76.52024-07
Meta Llama 3 8BNous Research76.52024-07
Phi 3 mini 4K instructMicrosoft75.82024-07
Yi 9B01.AI75.52024-07
Meta Llama 3 8B InstructMeta74.52024-07
Meta Llama 3 8B InstructNous Research74.52024-07
Yi 1.5 9B Chat01.AI74.52024-07
Mistral 7B v0.3Mistral AI74.22024-07
mistral 7B v0.3 bnb 4bitUnsloth74.22024-07
Hermes 2 Theta Llama 3 8BNous Research73.52024-07
Yi 6B01.AI71.62024-07
Yi 6B Chat01.AI71.62024-07
Hermes 2 Pro Llama 3 8BNous Research71.22024-07
Qwen2 1.5B InstructAlibaba / Qwen Team69.32024-07
Yi 1.5 6B01.AI69.32024-07
Mistral 7B Instruct v0.2Mistral AI68.62024-07
Yi 1.5 6B Chat01.AI68.02024-07
falcon 40BTII67.32024-07
phi 2Microsoft62.12024-07
chatglm2 6BZhipu AI53.32024-07
deepseek llm 7B baseDeepSeek52.32024-07
deepseek llm 7B chatDeepSeek52.32024-07
Qwen2 0.5B InstructAlibaba / Qwen Team51.62024-07
gemma 2BGoogle DeepMind46.72024-07
gemma 2B itGoogle DeepMind45.12024-07
deepseek coder 6.7B baseDeepSeek40.22024-07
deepseek coder 6.7B instructDeepSeek39.22024-07
deepseek coder 1.3B baseDeepSeek30.42024-07
deepseek coder 1.3B instructDeepSeek30.42024-07
OLMo 1B hfAllen AI29.72024-07

Data

This page as JSON · Edit on GitHub