Model

Gemini 1.5 Flash

Google

deepmind.google

Coding

Coding
BenchmarkScoreMetricHarnessSourceConfidenceLink
MMSQL65.8tdexN/AIndependentmediumarxiv.org

Multimodal

Multimodal
BenchmarkScoreMetricHarnessSourceConfidenceLink
OVBench50.7avgN/AIndependentmediumarxiv.org
Video-MME75.0accuracy_(%)N/AIndependentmediumarxiv.org

Reproduced by the community

Loading community runs…