MMLU: Virology

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

Also known as: virology

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 size166
Dataset licenceMIT
PublisherUC Berkeley

What it measures

The study of viruses: viral structure and replication, virus-caused disease, and the basics of immunology and vaccines as they relate to viral infection. 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
Yi 1.5 34B Chat01.AI62.72026-04
Yi 1.5 34B Chat 16K01.AI62.72024-07
Mixtral 8x22B Instruct v0.1Mistral AI59.02026-04
Yi 34B Chat01.AI58.42024-07
Yi 1.5 34B 32K01.AI57.82024-07
Nous Hermes 2 Yi 34BNous Research57.22024-07
Yi 1.5 9B Chat 16K01.AI57.22024-07
Yi 34B 200K01.AI57.22024-07
Meta Llama 3 70BMeta56.62024-07
Meta Llama 3 70B InstructMeta56.62026-04
Meta Llama 3 70B InstructNous Research56.62026-04
Yi 1.5 34B01.AI56.62026-04
Meta Llama 3 8BMeta56.02024-07
Meta Llama 3 8BNous Research56.02024-07
Nous Hermes 2 SOLAR 10.7BNous Research56.02024-07
Yi 1.5 9B Chat01.AI56.02024-07
Hermes 2 Theta Llama 3 8BNous Research54.82024-07
Yi 1.5 9B 32K01.AI54.22024-07
gemma 7B itGoogle DeepMind53.62024-07
Mistral 7B v0.3Mistral AI53.62024-07
mistral 7B v0.3 bnb 4bitUnsloth53.62024-07
Yi 1.5 9B01.AI53.62024-07
Hermes 2 Pro Llama 3 8BNous Research53.02024-07
Meta Llama 3 8B InstructMeta51.22024-07
Meta Llama 3 8B InstructNous Research51.22024-07
Mixtral 8x7B v0.1Mistral AI51.22026-04
Yi 9B01.AI51.22024-07
Mixtral 8x7B Instruct v0.1Mistral AI50.62026-04
Nous Hermes 2 Mixtral 8x7B DPONous Research50.62024-07
Yi 1.5 6B01.AI50.62024-07
Yi 1.5 6B Chat01.AI50.02024-07
Mistral 7B Instruct v0.2Mistral AI49.42024-07
Phi 3 mini 4K instructMicrosoft49.42024-07
deepseek llm 7B baseDeepSeek47.62024-07
deepseek llm 7B chatDeepSeek47.62024-07
falcon 40BTII47.62024-07
phi 2Microsoft47.02024-07
Phi 3 mini 128K instructMicrosoft47.02024-07
Qwen2 1.5B InstructAlibaba / Qwen Team46.42024-07
Yi 6B01.AI46.42024-07
Yi 6B Chat01.AI46.42024-07
gemma 2BGoogle DeepMind44.62024-07
chatglm2 6BZhipu AI41.62024-07
deepseek coder 6.7B baseDeepSeek41.02024-07
gemma 2B itGoogle DeepMind41.02024-07
deepseek coder 6.7B instructDeepSeek37.32024-07
Qwen2 0.5B InstructAlibaba / Qwen Team37.32024-07
deepseek coder 1.3B baseDeepSeek25.32024-07
deepseek coder 1.3B instructDeepSeek25.32024-07
OLMo 1B hfAllen AI22.92024-07

Data

This page as JSON · Edit on GitHub