Jacqkues/sev-generic-bench download history

Jacqkues/sev-generic-bench is a dataset on the Hugging Face Hub. In the last 30 days it was downloaded 40 times (12 in the last 7 days), and 113 times in total. It ranks #316,612 among datasets by monthly downloads.

Sev generic benchmark A generic evaluation of typed, calibrated vision decisions. It is built from five public VLM benchmarks: MMStar, MMBench-en dev (1,000 sampled items), MME, RealWorldQA and HallusionBench (image split). Every item is cast as a TypeSafe-style question: a choice ove

Models trained on sev-generic-bench

1 models list it as training data.

Open Jacqkues/sev-generic-bench on Hugging Face

Sister project: Paper Pulse, the upvote history of every Hugging Face Daily Paper.