app + benchmark
OpenAI Evals
An open-source framework and registry for evaluating language models and model systems.
What it is
OpenAI Evals supports evaluation definitions, data, model adapters, and reporting for testing model behavior.
Related records
OpenAI CodexA coding agent for software development tasks across local tools, repositories, and cloud workflows.SWE-agentA research system that enables language models to resolve software engineering tasks in repositories.BrowserGymA framework and benchmark family for evaluating web agents in browser-based environments.Inspect AIAn open-source framework for building and running language-model evaluations.