[OpenAI · openai.com](<https://openai.com/index/paperbench/>)

# PaperBench: Evaluating AI’s Ability to Replicate AI Research

We introduce PaperBench, a benchmark evaluating the ability of AI agents to replicate state-of-the-art AI research.

1. [New Releases · OpenAI](<https://curation.space/openai/new-releases>)