Your model / ????
XX%
Sign in to run your model of choice.
apgard Youth AI Safety Benchmark
See how frontier models perform on youth AI safety risks and run our benchmarks for your own use case.
Overview
Model results on youth mental well-being and youth sexual safety, built upon KORA's open-source AI child safety benchmark tool and extended with our expert-informed taxonomies. Models are rated by highest overall benchmark score; select a model to view its risk breakdown and scenario assessments.
Sign in to run the benchmarks on your own models here.