Benchmark

OpenBookQA

OpenBookQA is a question-answering dataset modeled after open book exams for assessing human understanding. It contains 5,957 multiple-choice…

Modality
text
Categories
reasoning, general
Openness
unknown
Source
llm_stats
Reported scores
5

Reported scores

llm_stats

ModelOrganizationReported valueReportedEvidence
Hermes 3 70BNous Research0.4942024-08-15evidence
Mistral NeMo InstructMistral0.6062024-07-18evidence
Phi 4 MiniMicrosoft0.7922025-02-01evidence
Phi-3.5-mini-instructMicrosoft0.7922024-08-23evidence
Phi-3.5-MoE-instructMicrosoft0.8962024-08-23evidence

Scores are partitioned by the source that reported them and are never merged into a single cross-source ranking, because the sources measure different things and say so.