{"url":"/dataset/oc20","name":"OC20","full_name":"Open Catalyst 2020","description_markdown":"**Open Catalyst 2020** is a dataset for catalysis in chemical engineering. Focusing on molecules that are important in renewable energy applications, the OC20 data set comprises over 1.3 million relaxations of molecular adsorptions onto surfaces, the largest data set of electrocatalyst structures to date.","description_withheld":null,"homepage":"https://github.com/Open-Catalyst-Project/ocp/blob/master/DATASET.md","introduced_date":"2020-10-20","introduced_date_note":null,"introduced_by":{"paper":"/paper/the-open-catalyst-2020-oc20-dataset-and","title":"The Open Catalyst 2020 (OC20) Dataset and Community Challenges","first_author":"Lowik Chanussot","url":null},"license":{"name":"Creative Commons Attribution 4.0 License","url":null},"modalities":[],"tasks":[{"name":"Initial Structure to Relaxed Energy (IS2RE)","url":"/task/initial-structure-to-relaxed-energy-is2re","datasets_with_task":"/datasets/task/initial-structure-to-relaxed-energy-is2re"},{"name":"Initial Structure to Relaxed Energy (IS2RE), Direct","url":"/task/initial-structure-to-relaxed-energy-is2re-1","datasets_with_task":"/datasets/task/initial-structure-to-relaxed-energy-is2re-1"}],"languages":[],"variants":["OC20"],"data_loaders":[{"repo":"https://github.com/Open-Catalyst-Project/ocp","url":"https://github.com/Open-Catalyst-Project/ocp","frameworks":["pytorch"]}],"num_papers_in_archive":77,"source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28"},"benchmarks":[{"leaderboard":"/sota/initial-structure-to-relaxed-energy-is2re-on","task":"Initial Structure to Relaxed Energy (IS2RE)","dataset_variant":"OC20","rows":4,"metrics":["Energy MAE"],"first_row_in_archive_order":{"model":"GemNet-OC","paper":"/paper/how-do-graph-networks-generalize-to-large-and","metrics":{"Energy MAE":"0.348"},"code_links":[]},"note":"rows are the archive's own order at snapshot; nothing here re-ranks them"}],"papers_with_a_benchmark_row":[{"paper":"/paper/how-do-graph-networks-generalize-to-large-and","title":"GemNet-OC: Developing Graph Neural Networks for Large and Diverse Molecular Simulation Datasets","date":"2022-04-06","rows_on_this_dataset":1,"code_links":0,"syntology":null},{"paper":"/paper/towards-training-billion-parameter-graph-1","title":"Towards Training Billion Parameter Graph Neural Networks for Atomic Simulations","date":"2022-03-18","rows_on_this_dataset":1,"code_links":1,"syntology":null},{"paper":"/paper/rotation-invariant-graph-neural-networks","title":"Rotation Invariant Graph Neural Networks using Spin Convolutions","date":"2021-06-17","rows_on_this_dataset":1,"code_links":1,"syntology":null},{"paper":"/paper/very-deep-graph-neural-networks-via-noise","title":"Simple GNN Regularisation for 3D Molecular Property Prediction & Beyond","date":"2021-06-15","rows_on_this_dataset":1,"code_links":1,"syntology":{"read_at":"2026-09-24T18:15:14+00:00","samples_harvested":1,"samples_ran":1,"samples_unverified":0,"pointer_only_for_licence":1,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim."}}],"syntology_totals":{"read_at":"2026-09-24T18:15:14+00:00","papers_with_samples":1,"samples_harvested":1,"samples_ran":1,"samples_unverified":0,"pointer_only_for_licence":1,"papers_with_no_sample_that_ran":0,"note":"the per-paper counts above, summed; not a rate"},"papers_note":"The archive never published its papers-using-dataset list; these are papers with a leaderboard row on this dataset's benchmarks."}