Open ai evaluation

OpenAI evaluation refers to `openai/evals`—an open-source framework toolkit hosted on GitHub that provides standardized system evaluation registries and benchmarking tool loops to test, quantify, and grade the performance of large language models.