OpenCSP: A Deep Learning Framework for Crystal Structure Prediction from Ambient to High Pressure

OpenCSP: A Deep Learning Framework for Crystal Structure Prediction from Ambient to High Pressure | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article OpenCSP: A Deep Learning Framework for Crystal Structure Prediction from Ambient to High Pressure Yinan Wang, Xiaoyang Wang, Zhenyu Wang, Jing Wu, Jian Lv, Weinan E, and 1 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7615850/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 10 You are reading this latest preprint version Abstract High-pressure crystal structure prediction (CSP) underpins advances in condensed matter physics, planetary science, and materials discovery. Yet, most large atomistic models are trained on near-ambient, equilibrium data, leading to degraded stress accuracy at tens to hundreds of gigapascalsand sparse coverage of pressure-stabilized stoichiometries and dense coordination motifs. Here,we introduce OpenCSP, a machine learning framework for CSP tasks spanning ambient to high-pressure conditions. This framework comprises an open-source pressure-resolved dataset alongsidea suite of publicly available atomistic models that are jointly optimized for accuracy in energy, force, and stress predictions. The dataset is constructed via randomized high-pressure sampling anditeratively refined through an uncertainty-guided concurrent learning strategy, which enriches under-represented compression regimes while suppressing redundant DFT labeling. Despite employing atraining corpus one to two orders of magnitude smaller than those of leading large models, OpenCSP achieves comparable or superior performance in high-pressure enthalpy ranking and stability prediction. Across benchmark CSP tasks spanning a wide pressure window, our models match or surpass MACE-MPA-0, MatterSim v1 5M, and GRACE-2L-OAM, with the largest gains observed at elevated pressures. These results demonstrate that targeted, pressure-aware data acquisition coupledwith scalable architectures enables data-efficient, high-fidelity CSP, paving the way for autonomous materials discovery under ambient and extreme conditions. Physical sciences/Chemistry Physical sciences/Materials science Full Text Additional Declarations No competing interests reported. Cite Share Download PDF Status: Under Review Version 1 posted Editorial decision: Revision requested 03 Mar, 2026 Reviews received at journal 22 Feb, 2026 Reviewers agreed at journal 06 Feb, 2026 Reviewers agreed at journal 05 Nov, 2025 Reviews received at journal 17 Oct, 2025 Reviewers agreed at journal 28 Sep, 2025 Reviewers invited by journal 26 Sep, 2025 Editor assigned by journal 18 Sep, 2025 Submission checks completed at journal 18 Sep, 2025 First submitted to journal 14 Sep, 2025 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-7615850","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":522038819,"identity":"81a002c2-927c-41cd-8824-7014613c0b8b","order_by":0,"name":"Yinan Wang","email":"","orcid":"","institution":"Artificial Intelligence for Science Institute","correspondingAuthor":false,"prefix":"","firstName":"Yinan","middleName":"","lastName":"Wang","suffix":""},{"id":522038821,"identity":"ee3fe6a3-31e3-455e-bbd3-0e3ef6bc2198","order_by":1,"name":"Xiaoyang Wang","email":"","orcid":"","institution":"Institute of Applied Physics and Computational Mathematics","correspondingAuthor":false,"prefix":"","firstName":"Xiaoyang","middleName":"","lastName":"Wang","suffix":""},{"id":522038822,"identity":"978322d3-3a59-4ef7-9d7c-209457a9d539","order_by":2,"name":"Zhenyu Wang","email":"","orcid":"","institution":"Jilin University","correspondingAuthor":false,"prefix":"","firstName":"Zhenyu","middleName":"","lastName":"Wang","suffix":""},{"id":522038823,"identity":"ad3350de-1477-47e0-ad89-1ed3a8f58f39","order_by":3,"name":"Jing Wu","email":"","orcid":"","institution":"Artificial Intelligence for Science Institute","correspondingAuthor":false,"prefix":"","firstName":"Jing","middleName":"","lastName":"Wu","suffix":""},{"id":522038824,"identity":"a0d3171c-a419-44f6-9ad6-12338794c963","order_by":4,"name":"Jian Lv","email":"","orcid":"","institution":"Jilin University","correspondingAuthor":false,"prefix":"","firstName":"Jian","middleName":"","lastName":"Lv","suffix":""},{"id":522038825,"identity":"5f0c7d3b-0740-4cad-a604-5511540a1d8b","order_by":5,"name":"Weinan E","email":"","orcid":"","institution":"Artificial Intelligence for Science Institute","correspondingAuthor":false,"prefix":"","firstName":"Weinan","middleName":"","lastName":"E","suffix":""},{"id":522038826,"identity":"e05a00cb-a56d-4681-82ea-8ad7ac45a7ab","order_by":6,"name":"Han Wang","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA2klEQVRIiWNgGAWjYLACxgYGBn5mkrVINpOsxeAAsaoNjp89/IJxh12e8XHmZw9+5jDI8zcAGXi1nMlLs2A8k1xsdpjN3LB3G4PhjANs5gb4tJgdyDEzYGxjTtx2mMFMmnEbA+MGBh42Cbxazr8BaalP3NzM/g2kxZ6wlhs5xg8Y2w4nbmDmAduSSFCL/Y03ZgyMbccTZxzmKQf6RSJ5xmE2M7xaJPtzjD8wtlUn9vcf3/bg5zYb2/725md4tQABm/QfKAOIgYqJSAbMH2B6CasdBaNgFIyCEQkArD1DdBh3sQYAAAAASUVORK5CYII=","orcid":"","institution":"Institute of Applied Physics and Computational Mathematics","correspondingAuthor":true,"prefix":"","firstName":"Han","middleName":"","lastName":"Wang","suffix":""}],"badges":[],"createdAt":"2025-09-15 03:38:21","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7615850/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7615850/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":93167366,"identity":"73474f81-d9d3-4e3d-8100-f4e33e46e775","added_by":"auto","created_at":"2025-10-09 18:17:49","extension":"json","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":8402,"visible":true,"origin":"","legend":"","description":"","filename":"fd922aa459554fefb9ca6fa85998d2af.json","url":"https://assets-eu.researchsquare.com/files/rs-7615850/v1/58e0ec9ec5d878e6050a7279.json"},{"id":93168189,"identity":"dae82a48-2314-42a8-a6f2-80f5e4cd5e17","added_by":"auto","created_at":"2025-10-09 18:25:52","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1737302,"visible":true,"origin":"","legend":"","description":"","filename":"OpenCSP.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7615850/v1_covered_b97dd4c2-8a56-45fd-b150-2b332e98028e.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"OpenCSP: A Deep Learning Framework for Crystal Structure Prediction from Ambient to High Pressure","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":true,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":true,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"npj-computational-materials","isNatureJournal":false,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"npjcompumats","sideBox":"Learn more about [npj Computational Materials](http://www.nature.com/npjcompumats/)","snPcode":"41524","submissionUrl":"https://mts-npjcompumats.nature.com/","title":"npj Computational Materials","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"ejp","reportingPortfolio":"NPJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-7615850/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7615850/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"High-pressure crystal structure prediction (CSP) underpins advances in condensed matter physics, planetary science, and materials discovery. Yet, most large atomistic models are trained on near-ambient, equilibrium data, leading to degraded stress accuracy at tens to hundreds of gigapascalsand sparse coverage of pressure-stabilized stoichiometries and dense coordination motifs. Here,we introduce OpenCSP, a machine learning framework for CSP tasks spanning ambient to high-pressure conditions. This framework comprises an open-source pressure-resolved dataset alongsidea suite of publicly available atomistic models that are jointly optimized for accuracy in energy, force, and stress predictions. The dataset is constructed via randomized high-pressure sampling anditeratively refined through an uncertainty-guided concurrent learning strategy, which enriches under-represented compression regimes while suppressing redundant DFT labeling. Despite employing atraining corpus one to two orders of magnitude smaller than those of leading large models, OpenCSP achieves comparable or superior performance in high-pressure enthalpy ranking and stability prediction. Across benchmark CSP tasks spanning a wide pressure window, our models match or surpass MACE-MPA-0, MatterSim v1 5M, and GRACE-2L-OAM, with the largest gains observed at elevated pressures. These results demonstrate that targeted, pressure-aware data acquisition coupledwith scalable architectures enables data-efficient, high-fidelity CSP, paving the way for autonomous materials discovery under ambient and extreme conditions.","manuscriptTitle":"OpenCSP: A Deep Learning Framework for Crystal Structure Prediction from Ambient to High Pressure","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-10-09 18:17:45","doi":"10.21203/rs.3.rs-7615850/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2026-03-04T00:15:34+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-02-22T23:52:15+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"284025912160508702705500736688416531843","date":"2026-02-06T10:01:53+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"195662822197880280931876543464112071824","date":"2025-11-05T08:29:05+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-10-17T15:52:21+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"294473774304501365063443712994567633113","date":"2025-09-28T09:28:18+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-09-26T07:14:43+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-09-18T18:07:42+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-09-18T05:15:18+00:00","index":"","fulltext":""},{"type":"submitted","content":"npj Computational Materials","date":"2025-09-15T03:35:17+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"npj-computational-materials","isNatureJournal":false,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"npjcompumats","sideBox":"Learn more about [npj Computational Materials](http://www.nature.com/npjcompumats/)","snPcode":"41524","submissionUrl":"https://mts-npjcompumats.nature.com/","title":"npj Computational Materials","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"ejp","reportingPortfolio":"NPJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"171bf724-9dfe-49c5-ab0f-2fe553379812","owner":[],"postedDate":"October 9th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[{"id":55481231,"name":"Physical sciences/Chemistry"},{"id":55481232,"name":"Physical sciences/Materials science"}],"tags":[],"updatedAt":"2026-05-21T07:40:08+00:00","versionOfRecord":[],"versionCreatedAt":"2025-10-09 18:17:45","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7615850","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7615850","identity":"rs-7615850","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}