Papers with Code

A free website (2018–2025) that linked machine learning papers, mostly from arXivarXiv, prop. n.: A free website where scientists post research papers before, or instead of, formal publication. Most new AI research appears there first., to open-source implementations and organized them by task, dataset and method, with community-maintained 'state of the artstate of the art, n.: The best performance or most advanced technique currently known in a field. In AI papers, a state-of-the-art result is one that beats every earlier published result on a test.' leaderboardsleaderboard, n.: A public ranking table showing which AI systems score best on a test, updated as new results arrive — like a sports league table for models. recording the best reported result on thousands of benchmarksbenchmark, n.: A fixed, shared test — a set of questions or tasks with an agreed way of scoring — used to compare AI systems with one another and to track progress over time.. For several years it was the default place to check which method led on a benchmark and to find code, and its taxonomy and data were widely used in studies of scientific progress. Critics argued that its leaderboards encouraged chasing small benchmark gains, and many entries were unverified or out of date. Meta retired it in July 2025, pointing users toward Hugging FaceHugging Face, prop. n.: A company whose website is the main public library for sharing AI models and datasets — often called the GitHub of machine learning — and which makes widely used free AI software.'s paper pages.