Benchmark

ScienceQA Visual Benchmark

What is ScienceQA Visual?

ScienceQA Visual is a multimodal science question answering benchmark consisting of 21,208 multiple-choice questions from elementary and high school science…

Released
2022-09-20
Modality
multimodal
Evaluates
multimodal, reasoning, vision
Openness
unknown
Importer
LLM Stats
Reported scores
1

Source provenance

ScienceQA Visual results and reported scores

LLM Stats

ModelOrganizationReported valueReportedEvidence
Phi-4-multimodal-instructMicrosoft0.9752025-02-01evidence

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.

Which sources cite ScienceQA Visual?

Related multimodal, reasoning, vision benchmarks