Model

Gemini Pro Vision

Google

deepmind.google

Multimodal

Multimodal
BenchmarkScoreMetricHarnessSourceConfidenceLink
MM-Vet v257.2gpt-4_scoreN/AIndependentmediumarxiv.org

Safety & Alignment

Safety & Alignment
BenchmarkScoreMetricHarnessSourceConfidenceLink
AutoHallusion51.4overall_accuracyN/AIndependentmediumpaperswithcode.com

Reproduced by the community

Loading community runs…