Open Graph Benchmark | A collection of benchmark datasets, data-loaders and evaluators for graph machine learning in PyTorch.

<!DOCTYPE html> <html lang="en-US">  <script async src="https://www.googletagmanager.com/gtag/js?id=UA-165257358-1"></script> <script> window.dataLayer = window.dataLayer || []; function gtag(){dataLayer.push(arguments);} gtag('js', new Date()); gtag('config', 'UA-165257358-1'); </script> <head> <meta charset="utf-8"> <meta http-equiv="X-UA-Compatible" content="IE=edge"> <meta name="viewport" content="width=device-width, initial-scale=1"> <link rel="stylesheet" href="/assets/css/main.css"> <link rel="stylesheet" href="/assets/css/font-awesome.min.css"> <link rel="shortcut icon" href="/favicon.ico?1">  <title>Open Graph Benchmark | A collection of benchmark datasets, data-loaders and evaluators for graph machine learning in PyTorch.</title> <meta name="generator" content="Jekyll v3.10.0" /> <meta property="og:title" content="Open Graph Benchmark" /> <meta property="og:locale" content="en_US" /> <meta name="description" content="A collection of benchmark datasets, data-loaders and evaluators for graph machine learning in PyTorch." /> <meta property="og:description" content="A collection of benchmark datasets, data-loaders and evaluators for graph machine learning in PyTorch." /> <link rel="canonical" href="https://snap-stanford.github.io/ogb-web/" /> <meta property="og:url" content="https://snap-stanford.github.io/ogb-web/" /> <meta property="og:site_name" content="Open Graph Benchmark" /> <meta property="og:type" content="website" /> <meta name="twitter:card" content="summary" /> <meta property="twitter:title" content="Open Graph Benchmark" /> <script type="application/ld+json"> {"@context":"https://schema.org","@type":"WebSite","description":"A collection of benchmark datasets, data-loaders and evaluators for graph machine learning in PyTorch.","headline":"Open Graph Benchmark","name":"Open Graph Benchmark","url":"https://snap-stanford.github.io/ogb-web/"}</script>  <link rel="alternate" type="application/rss+xml" title="Open Graph Benchmark" href="https://snap-stanford.github.io/ogb-web/feed.xml" /> </head> <body> <nav class="navbar navbar-default navbar-fixed-top"> <div class="container navbar-container"> <div class="navbar-header"> <button type="button" class="navbar-toggle collapsed" data-toggle="collapse" data-target="#navbar" aria-expanded="false" aria-controls="navbar"> <span class="sr-only">Toggle navigation</span> <span class="icon-bar"></span> <span class="icon-bar"></span> <span class="icon-bar"></span> </button> <a class="navbar-brand" href="/"> <span><img src="/assets/img/OGB_pure.png" alt="Logo"></span> </a> </div> <div id="navbar" class="collapse navbar-collapse"> <div class="navbar-right"> <ul class="nav navbar-nav"> <li ><a href="/docs/home/">Get Started</a></li> <li ><a href="/docs/update/">Updates</a></li> <li ><a class="dropdown-toggle" data-toggle="dropdown" href="#"> Large-Scale Challenge <span class="caret"></span> </a> <ul class="dropdown-menu"> <li><a href="/docs/lsc/">Overview</a></li> <li class="divider"></li> <li><a href="/docs/lsc/mag240m/">MAG240M</a></li> <li class="divider"></li> <li><a href="/docs/lsc/wikikg90mv2/">WikiKG90Mv2</a></li> <li class="divider"></li> <li><a href="/docs/lsc/pcqm4mv2/">PCQM4Mv2</a></li> <li class="divider"></li> <li><a href="/docs/lsc/rules/">Rules</a></li> <li class="divider"></li> <li><a href="/docs/lsc/leaderboards/">Leaderboards</a></li> <li class="divider"></li> <li><a href="https://arxiv.org/abs/2103.09430">LSC Paper</a></li> <li class="divider"></li> <li><a href="/neurips2022/">2nd LSC@NeurIPS 2022</a></li> <li class="divider"></li> <li><a href="/kddcup2021/">1st LSC@KDD Cup 2021 (archive)</a></li> </ul> </li> <li ><a class="dropdown-toggle" data-toggle="dropdown" href="#"> Datasets <span class="caret"></span> </a> <ul class="dropdown-menu"> <li><a href="/docs/dataset_overview/">Overview</a></li> <li class="divider"></li> <li><a href="/docs/nodeprop/">Node Property Prediction</a></li> <li class="divider"></li> <li><a href="/docs/linkprop/">Link Property Prediction</a></li> <li class="divider"></li> <li><a href="/docs/graphprop/">Graph Property Prediction</a></li> <li class="divider"></li> <li><a href="/docs/lsc/">Large-Scale Challenge</a></li> </ul> </li> <li ><a class="dropdown-toggle" data-toggle="dropdown" href="#"> Leaderboards <span class="caret"></span> </a> <ul class="dropdown-menu"> <li><a href="/docs/leader_overview/">Overview</a></li> <li class="divider"></li> <li><a href="/docs/leader_rules/">Rules</a></li> <li class="divider"></li> <li><a href="/docs/leader_nodeprop/">Node Property Prediction</a></li> <li class="divider"></li> <li><a href="/docs/leader_linkprop/">Link Property Prediction</a></li> <li class="divider"></li> <li><a href="/docs/leader_graphprop/">Graph Property Prediction</a></li> <li class="divider"></li> <li><a href="/docs/lsc/leaderboards/">Large-Scale Challenge</a></li> </ul> </li> <li ><a class="dropdown-toggle" data-toggle="dropdown" href="#"> Papers <span class="caret"></span> </a> <ul class="dropdown-menu"> <li><a href="https://arxiv.org/abs/2005.00687">OGB paper</a></li> <li class="divider"></li> <li><a href="https://arxiv.org/abs/2103.09430">OGB-LSC paper</a></li> </ul> </li> <li ><a href="/docs/team/">Team</a></li> <li ><a href="https://github.com/snap-stanford/ogb">Github</a></li> </div> </div> </div> </nav> <div class="page-content"> <div class="wrapper"> <div class="header-container jumbotron"> <div class="container"> <h1>Open Graph Benchmark</h1> <p><strong>Benchmark datasets, data loaders and evaluators for graph machine learning<strong></p> <p><a class="btn btn-primary btn-lg" href="/docs/home/" role="button">Get Started</a> <a class="btn btn-warning btn-lg" href="/docs/lsc/" role="button">OGB-LSC</a> </p>  </div> </div> <div class="container"> <div class="row"> <div class="col-md-6"> <p class="lead"> <font size="5"> The Open Graph Benchmark (OGB) is a collection of realistic, large-scale, and diverse benchmark datasets for machine learning on graphs. OGB datasets are automatically downloaded, processed, and split using the <a href="/docs/home/#dataloader">OGB Data Loader</a>. The model performance can be evaluated using the <a href="/docs/home/#evaluator">OGB Evaluator</a> in a unified manner. <br/> OGB is a community-driven initiative in active development. We expect the benchmark datasets to evolve. To keep up to date to major updates, <b>subscribe to our google group <a href="https://groups.google.com/forum/#!forum/open-graph-benchmark">here</a></b>. </font> </p> </div> <div class="col-md-6 text-center"> <img src="/assets/img/OGB.png" alt="Jekyll logo" class="img-responsive"> </div> </div> <hr> <div class="row"> <div class="col-sm-4"> <h1 class="text-center"><a href="/docs/dataset_overview/"><i class="fa fa-database" aria-hidden="true"></i></a></h1> <h3 class="text-center"><a href="/docs/dataset_overview/">Realistic datasets</a></h3> <p class="f3-light">OGB provides a diverse set of challenging and realistic <a href="/docs/dataset_overview/">benchmark datasets</a> that are of varying sizes and cover a variety graph machine learning tasks, including prediction of <a href="/docs/nodeprop/">node, </a><a href="/docs/linkprop/"> link, </a> and <a href="/docs/graphprop/">graph</a> properties. </p> </div> <div class="col-sm-4"> <h1 class="text-center"><a href="/docs/home/#dataloader"><i class="fa fa-cogs" aria-hidden="true"></i></a></h1> <h3 class="text-center"><a href="/docs/home/#dataloader">Flexible data loaders</a></h3> <p class="f3-light">OGB fully automates dataset processing. The OGB data loaders automatically download and process graphs, provide graph objects that are fully compatible with <a href="https://pytorch-geometric.readthedocs.io/">Pytorch Geometric</a> and <a href="https://www.dgl.ai/">DGL</a>. </p> </div> <div class="col-sm-4"> <h1 class="text-center"><a href="/docs/home/#evaluator"><i class="fa fa-bar-chart" aria-hidden="true"></i></a></h1> <h3 class="text-center"><a href="/docs/home/#evaluator">Unified evaluation</a></h3> <p class="f3-light">OGB provides standardized dataset splits and evaluators that allow for easy and reliable comparison of different models in a unified manner. OGB uses <a href="/docs/leader_overview/">leaderboards</a> to keep track of the state-of-the-art. </p> </div> </div> <hr> <div class="col-md-12 text-center"> <a href="https://arxiv.org/abs/2005.00687"> <img src="/assets/img/paper_download.png" alt="Jekyll logo" class="img-responsive"> </a> </div> <p style="color:#FFFFFF">a</p> <div class="col-md-12 text-center"> <a href="https://arxiv.org/abs/2103.09430"> <img src="/assets/img/paper_download_lsc.png" alt="Jekyll logo" class="img-responsive"> </a> </div> </div> </div> <footer class="footer"> <div class="container"> <p class="text-center"> Open Graph Benchmark 2025 | Powered by <a href="https://github.com/aksakalli/jekyll-doc-theme">Jekyll Doc Theme</a> </p>  </div> </footer> <script> var baseurl = '' </script> <script src="https://code.jquery.com/jquery-1.12.4.min.js"></script> <script src="/assets/js/bootstrap.min.js "></script> <script src="/assets/js/typeahead.bundle.min.js "></script> <script src="/assets/js/main.js "></script> </body> </html>

CINXE.COM

Open Graph Benchmark | A collection of benchmark datasets, data-loaders and evaluators for graph machine learning in PyTorch.