GenAI Platform Evaluator documentation
Compare AI platform profiles and capabilities in a shared evaluation workspace.
Compare a shared set of evidence
The prototype loads model and capability tables and supports adding JSON evaluation records to a session. Comparison views and charts help explore differences across the supplied profiles.
Interpret the comparison
Results reflect the information in the imported profiles. Check the date, source, and scoring criteria behind each record before drawing conclusions about a platform's current capabilities. Category normalization can help compare chart shapes, but it also affects how values are displayed.
Current availability
GenAI Platform Evaluator is a working prototype. Contact NVAA to discuss a walkthrough and the evaluation data you want to bring together.