Model

LLaVA

Vendor N/A

Knowledge & Reasoning

Knowledge & Reasoning
BenchmarkScoreMetricHarnessSourceConfidenceLink
ScienceQA92.5scoreN/AIndependentmediumscienceqa.github.io

Multimodal

Multimodal
BenchmarkScoreMetricHarnessSourceConfidenceLink
MVBench36.0avg.N/AIndependentmediumarxiv.org
Winoground24.8text_scoreN/AIndependentmediumarxiv.org

Reproduced by the community

Loading community runs…