Benchmark Radar™
Star —

VIPER Benchmark

Explore score history ↗

VIPER evaluates vision-language models on toxicologic pathology via 1,251 questions over 419 H&E-stained rat histology images across organ systems, covering multiple-choice, KPrim, and free-text formats with domain-expert validation.

Released
2026-08-26
Institutions
University of Bern · University of California, Davis · University of Augsburg
Evaluates
Biology & Drug Discovery, Multimodal
Openness
unknown
Importer
Claire Radar
Review status
ai-reviewed

VIPER paper and code

No reported scores are on record for this benchmark yet.

Which sources cite VIPER?

Related Biology & Drug Discovery, Multimodal benchmarks