{"url":"/dataset/geom-drugs","name":"GEOM-DRUGS","full_name":null,"description_markdown":"GEOM-DRUGS is a dataset of 430,000 large organic molecules of up to 180 atoms from [Axelrod and Gómez-Bombarelli, Nature Scientific Data, 2022](https://www.nature.com/articles/s41597-022-01288-4). \r\n\r\nThe dataset used by most machine learning papers is a processed version from: [link](https://github.com/cvignac/MiDi) or [link](https://huggingface.co/datasets/chaitjo/GEOM-DRUGS_ADiT). \r\n\r\nThe original, unprocessed dataset is available: [link](https://github.com/learningmatter-mit/geom).","description_withheld":null,"homepage":"https://github.com/learningmatter-mit/geom","introduced_date":"2023-02-17","introduced_date_note":null,"introduced_by":{"paper":"/paper/midi-mixed-graph-and-3d-denoising-diffusion","title":"MiDi: Mixed Graph and 3D Denoising Diffusion for Molecule Generation","first_author":"Clement Vignac","url":null},"license":null,"modalities":[{"name":"Graphs","url":"/datasets/modality/graphs"},{"name":"3D","url":"/datasets/modality/3d"}],"tasks":[{"name":"Unconditional Molecule Generation","url":"/task/unconditional-molecule-generation","datasets_with_task":"/datasets/task/unconditional-molecule-generation"}],"languages":[],"variants":["GEOM-DRUGS"],"data_loaders":[],"num_papers_in_archive":20,"source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28"},"benchmarks":[{"leaderboard":"/sota/unconditional-molecule-generation-on-geom","task":"Unconditional Molecule Generation","dataset_variant":"GEOM-DRUGS","rows":5,"metrics":["PoseBusters Validity","Validity","PoseBusters Atoms Connected"],"first_row_in_archive_order":{"model":"TABASCO","paper":"/paper/tabasco-a-fast-simplified-model-for-molecular-1","metrics":{"PoseBusters Validity":"92","Validity":"97"},"code_links":[{"title":"carlosinator/tabasco","url":"https://github.com/carlosinator/tabasco"}]},"note":"rows are the archive's own order at snapshot; nothing here re-ranks them"}],"papers_with_a_benchmark_row":[{"paper":"/paper/tabasco-a-fast-simplified-model-for-molecular-1","title":"TABASCO: A Fast, Simplified Model for Molecular Generation with Improved Physical Quality","date":"2025-07-01","rows_on_this_dataset":1,"code_links":1,"syntology":null},{"paper":"/paper/all-atom-diffusion-transformers-unified","title":"All-atom Diffusion Transformers: Unified generative modelling of molecules and materials","date":"2025-03-05","rows_on_this_dataset":1,"code_links":1,"syntology":{"read_at":"2026-09-24T18:15:14+00:00","samples_harvested":7,"samples_ran":6,"samples_unverified":1,"pointer_only_for_licence":7,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim."}},{"paper":"/paper/efficient-3d-molecular-generation-with-flow","title":"SemlaFlow -- Efficient 3D Molecular Generation with Latent Attention and Equivariant Flow Matching","date":"2024-06-11","rows_on_this_dataset":1,"code_links":0,"syntology":null},{"paper":"/paper/navigating-the-design-space-of-equivariant","title":"Navigating the Design Space of Equivariant Diffusion-Based Generative Models for De Novo 3D Molecule Generation","date":"2023-09-29","rows_on_this_dataset":1,"code_links":0,"syntology":null},{"paper":"/paper/midi-mixed-graph-and-3d-denoising-diffusion","title":"MiDi: Mixed Graph and 3D Denoising Diffusion for Molecule Generation","date":"2023-02-17","rows_on_this_dataset":1,"code_links":1,"syntology":{"read_at":"2026-09-24T18:15:14+00:00","samples_harvested":13,"samples_ran":2,"samples_unverified":11,"pointer_only_for_licence":0,"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":2,"samples_harvested":20,"samples_ran":8,"samples_unverified":12,"pointer_only_for_licence":7,"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."}