all the models — AI benchmark observatory
← Benchmarks

ScienceQA

ScienceQA is the first large-scale multimodal science question answering benchmark with 21,208 multiple-choice questions covering 3 subjects (natural science, language science, social science), 26 topics, 127 categories, and 379 skills. The benchmark includes both text and image modalities, featuring detailed explanations and Chain-of-Thought reasoning to diagnose multi-hop reasoning ability.

id scienceqa · max 1 · 1 models reported

#ModelScore
1Phi-3.5-vision-instruct
Microsoft · open
0.91
ScienceQA Leaderboard · all the models