Evals is a framework for evaluating OpenAI models and an open-source registry of benchmarks.
Latest commits.
Builders behind this project.
