Introducing SimpleQA

OpenAI has released SimpleQA, a benchmark designed to measure the factual accuracy of language models on short, fact-seeking questions. The benchmark spans diverse topics and provides a standardized method to evaluate how well models avoid incorrect or hallucinated information. This development addresses growing concerns about reliability in AI-generated content, especially for workflows that depend on verified facts. For developers building AI-powered applications—such as research assistants, customer support systems, or content verification tools—SimpleQA offers a practical way to compare model factuality. By adopting this benchmark, builders can make informed decisions about model selection, identify areas for improvement through fine-tuning, and track progress over time. While SimpleQA is not a product itself, it directly influences the quality of widely used tools like ChatGPT, Claude, and Perplexity. As the industry prioritizes factual reliability, benchmarks like SimpleQA become essential for maintaining user trust and ensuring accuracy in automated systems.