[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-126480-en":3,"doc-seo-126480-105":31,"detail-sidebar-cat-0-en-105":93},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":20,"is_deleted":4,"is_public":21,"is_downloadable":21,"audit_status":21,"page_count":22,"language":23,"language_code":24,"site_id":25,"html_lang":24,"table_of_contents":26,"faqs":27,"seo_title":28,"seo_description":14,"update_tm":29,"read_time":30},126480,962084925290,"Ophelia","https://ap-avatar.wpscdn.com/davatar_085a072bc5b1113ac321206ff7593b45",8,"Research & Report","metatensor and metatomic - Foundational libraries for interoperable atomistic machine learning","Machine learning techniques are widely used to improve atomic-scale simulations, but progress is hindered by mismatched mathematical foundations and fragmented software ecosystems. To enable more reliable adoption of ML in atomistic workflows, the article presents two interoperable libraries: metatensor for portable storage and manipulation of arrays with sparse indices, metadata, and support for geometric information and gradients, and metatomic for portable model and model-metadata representation. Together they bridge data exchange between Python-based ML tools and established simulation codes such as Fortran, C, and C++, supporting training and distribution across simulation packages.","RESEARCH ARTICLE | FEBRUARY 11 2026  \nmetatensor and metatomic: Foundational libraries for interoperable atomistic machine learning   \nFilippo Bigi  ; Joseph W. Abbott  ; Philip Loche  ; Arslan Mazitov  ; Davide Tisi  ; Marcel F. Langer  ; Alexander Goscinski  ; Paolo Pegolo  ; Sanggyu Chong  ; Rohit Goswami  ; Pol Febrer  ;  \nSofiia Chorna  ; Matthias Kellner  ; Michele Ceriotti  ; Guillaume Fraux 􀀧   \nJ. Chem. Phys. 164, 064113 (2026)  \n[https://doi.org/10.1063/5.0304911](https://doi.org/10.1063/5.0304911)  \n􀀭  \nView Online  \n􀀱  \nExport Citation  \nArticles You May Be Interested In  \nRepresenting spherical tensors with scalar-based machine-learning models  \nJ. Chem. Phys. (October 2025)  \nFast and flexible long-range models for atomistic machine learning  \nJ. Chem. Phys. (April 2025)  \nVelocity updates from machine learning for stable molecular dynamics integration AIP Advances (January 2026)  \nThe Journal  \nof Chemical Physics  \nARTICLE  \n[pubs.aip.org/aip/jcp](pubs.aip.org/aip/jcp)  \nmetatensor and metatomic: Foundational libraries for interoperable atomistic machine learning   \n\n| Cite as: J. Chem. Phys. 164, 064113 (2026); doi: 10. 1063/5.0304911 Submitted: 1 October 2025 • Accepted: 20 January 2026 •\u003Cbr>Published Online: 11 February 2026 |  |  |  |\n| --- | --- | --- | --- |\n| Filippo Bigi,  Joseph W. Abbott,  Philip Loche,  Arslan Mazitov,  Davide Tisi,  Marcel F. Langer,  Alexander Goscinski,  Paolo Pegolo,  Sanggyu Chong,  Rohit Goswami,  Pol Febrer, \u003Cbr>Sofiia Chorna,  Matthias Kellner,  Michele Ceriotti,  and Guillaume Frauxa)  |  |  |  |\n| AFFILIATIONS\u003Cbr>Laboratory of Computational Science and Modeling, Institute of Materials, École Polytechnique Fédérale de Lausanne, 1015 Lausanne, Switzerland\u003Cbr>a)[Author to whom correspondence should be addressed:](Author to whom correspondence should be addressed: guillaume.fraux@epfl.ch)[ guillaume.fraux@epfl.ch](Author to whom correspondence should be addressed: guillaume.fraux@epfl.ch) |  |  |  |\n| ABSTRACT\u003Cbr>Incorporation of machine learning (ML) techniques into atomic-scale modeling has proven to be an extremely effective strategy to improve the accuracy and reduce the computational cost of simulations. It also entails conceptual and practical challenges, as it involves combining very different mathematical foundations as well as software ecosystems that are very well developed in their own right but do not share many commonalities. To address these issues and facilitate the adoption of ML in atomistic simulations, we introduce two dedicated software libraries. The first one, metatensor, provides multi-platform and multi-language storage and manipulation of arrays with many potentially sparse indices, designed from the ground up for atomistic ML applications. By combining the actual values with metadata that describes their nature and that facilitates the handling of geometric information and gradients with respect to the atomic positions, metatensor provides a common framework to enable data sharing between ML software—typically written in Python—and established atomistic modeling tools—typically written in Fortran, C, or C++ . The second library, metatomic, provides an interface to store an atomistic ML model and metadata about this model in a portable way, facilitating the implementation, training, and distribution of models, and their use across different simulation packages. We showcase a growing ecosystem of tools, including low-level libraries, training utilities, and interfaces with existing software packages, that demonstrate the effectiveness of metatensor and metatomic in bridging the gap between traditional simulation software and modern ML frameworks.\u003Cbr>© 2026 Author(s). All article content, except where otherwise noted, is licensed under a Creative Commons Attribution (CC BY) license ([https://creativecommons.org/licenses/by/4.0/](https://creativecommons.org/licenses/by/4.0/)). [https://doi.org/10.1063/5.0304911](https://doi.org/10.1063/5.0304911) |  |  |  ","cbCaiqEFHk1NFYM6","https://ap.wps.com/l/cbCaiqEFHk1NFYM6","pdf",7689918,7,1,20,"English","en",105,"# Motivation\n## Challenges from fragmented ML and simulation ecosystems\n# Proposed Solution\n## metatensor: portable array storage with metadata and gradients\n## metatomic: portable model and metadata for training and distribution\n# Results and Ecosystem\n## Tooling components that connect simulation software and ML frameworks","[{\"question\":\"What problem do metatensor and metatomic address in atomistic machine learning?\",\"answer\":\"They address conceptual and practical difficulties caused by combining different mathematical foundations and fragmented software ecosystems across ML and atomistic simulations.\"},{\"question\":\"What does metatensor provide?\",\"answer\":\"Metatensor provides a multi-platform, multi-language framework for storing and manipulating arrays with sparse indices, enriched with metadata for geometric information and gradients.\"},{\"question\":\"How does metatomic improve interoperability of ML models across simulation packages?\",\"answer\":\"Metatomic provides a portable way to store an atomistic ML model plus its metadata, enabling implementation, training, distribution, and use across different simulation packages.\"}]","metatensor and metatomic - Foundational libraries for interoperable atomistic machine learning | PDF",1785905286,50,{"code":4,"msg":32,"data":33},"ok",{"site_id":25,"language":24,"slug":34,"title":13,"keywords":35,"description":14,"schema_data":36,"social_meta":88,"head_meta":90,"extra_data":92,"updated_unix":29},"metatensor-and-metatomic-foundational-libraries-for-interoperable-atomistic-machine-learning","",{"@graph":37,"@context":87},[38,55,70],{"@type":39,"itemListElement":40},"BreadcrumbList",[41,45,49,52],{"item":42,"name":43,"@type":44,"position":21},"https://docshare.wps.com","Home","ListItem",{"item":46,"name":47,"@type":44,"position":48},"https://docshare.wps.com/document/","Document",2,{"item":50,"name":12,"@type":44,"position":51},"https://docshare.wps.com/document/research-report/",3,{"item":53,"name":13,"@type":44,"position":54},"https://docshare.wps.com/document/metatensor-and-metatomic-foundational-libraries-for-interoperable-atomistic-machine-learning/126480/",4,{"url":53,"name":13,"@type":56,"author":57,"headline":13,"publisher":59,"fileFormat":62,"inLanguage":24,"description":14,"dateModified":63,"datePublished":64,"encodingFormat":62,"isAccessibleForFree":65,"interactionStatistic":66},"DigitalDocument",{"name":9,"@type":58},"Person",{"url":42,"name":60,"@type":61},"DocShare","Organization","application/pdf","2026-08-22","2026-08-05",true,{"@type":67,"interactionType":68,"userInteractionCount":20},"InteractionCounter",{"@type":69},"ViewAction",{"@type":71,"mainEntity":72},"FAQPage",[73,79,83],{"name":74,"@type":75,"acceptedAnswer":76},"What problem do metatensor and metatomic address in atomistic machine learning?","Question",{"text":77,"@type":78},"They address conceptual and practical difficulties caused by combining different mathematical foundations and fragmented software ecosystems across ML and atomistic simulations.","Answer",{"name":80,"@type":75,"acceptedAnswer":81},"What does metatensor provide?",{"text":82,"@type":78},"Metatensor provides a multi-platform, multi-language framework for storing and manipulating arrays with sparse indices, enriched with metadata for geometric information and gradients.",{"name":84,"@type":75,"acceptedAnswer":85},"How does metatomic improve interoperability of ML models across simulation packages?",{"text":86,"@type":78},"Metatomic provides a portable way to store an atomistic ML model plus its metadata, enabling implementation, training, distribution, and use across different simulation packages.","https://schema.org",{"og:url":53,"og:type":89,"og:title":13,"og:site_name":60,"og:description":14},"article",{"robots":91,"canonical":53},"index,follow",{"doc_id":7,"site_id":25},{"code":4,"msg":5,"data":94},[95,99,103,107,112,116,120,123,127,130,134],{"id":21,"doc_module":4,"doc_module_name":47,"category_name":96,"show_sort_weight":97,"slug":98},"Story & Novel",90,"story-novel",{"id":48,"doc_module":4,"doc_module_name":47,"category_name":100,"show_sort_weight":101,"slug":102},"Literature",80,"literature",{"id":54,"doc_module":4,"doc_module_name":47,"category_name":104,"show_sort_weight":105,"slug":106},"Exam",70,"exam",{"id":108,"doc_module":4,"doc_module_name":47,"category_name":109,"show_sort_weight":110,"slug":111},5,"Comic",60,"comic",{"id":113,"doc_module":4,"doc_module_name":47,"category_name":114,"show_sort_weight":30,"slug":115},6,"Technology","technology",{"id":20,"doc_module":4,"doc_module_name":47,"category_name":117,"show_sort_weight":118,"slug":119},"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":47,"category_name":12,"show_sort_weight":121,"slug":122},30,"research-report",{"id":124,"doc_module":4,"doc_module_name":47,"category_name":125,"show_sort_weight":22,"slug":126},9,"Religion & Spirituality","religion-spirituality",{"id":22,"doc_module":4,"doc_module_name":47,"category_name":128,"show_sort_weight":22,"slug":129},"World Cup","world-cup",{"id":131,"doc_module":4,"doc_module_name":47,"category_name":132,"show_sort_weight":131,"slug":133},10,"Lifestyle","lifestyle",{"id":135,"doc_module":4,"doc_module_name":47,"category_name":136,"show_sort_weight":108,"slug":137},19,"General","general"]