{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,8]],"date-time":"2026-10-08T23:41:50Z","timestamp":1791502910121,"version":"4.3.4"},"reference-count":45,"publisher":"Springer Science and Business Media LLC","issue":"1","license":[{"start":{"date-parts":[[2026,8,21]],"date-time":"2026-08-21T00:00:00Z","timestamp":1787270400000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2026,9,21]],"date-time":"2026-09-21T00:00:00Z","timestamp":1789948800000},"content-version":"vor","delay-in-days":31,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"funder":[{"name":"Microsoft Research"}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["Nat Commun"],"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:p>Reliable description of bond breaking remains a major challenge for quantum chemistry due to the multireference character of the electronic structure in dissociating species. Multireference methods in particular suffer from large computational cost, which under the normal paradigm has to be paid anew for each system at a full price, ignoring commonalities in electronic structure across molecules. Quantum Monte Carlo with deep neural networks uniquely offers to exploit such commonalities by pretraining transferable wavefunction models, but all such attempts were so far limited in scope. Here, we bring this paradigm to fruition with Orbformer, a transferable wavefunction model pretrained on 22,000 equilibrium and dissociating structures that can be fine-tuned on unseen molecules reaching an accuracy\u2013cost ratio rivalling classical multireference methods. On established benchmarks as well as more challenging bond dissociations and Diels\u2013Alder reactions, Orbformer is the only method that consistently converges to chemical accuracy (1 kcal\/mol). This work turns the idea of amortizing the cost of solving the Schr\u00f6dinger equation over many molecules into a practical approach in quantum chemistry.<\/jats:p>","DOI":"10.1038\/s41467-026-76604-2","type":"journal-article","created":{"date-parts":[[2026,8,21]],"date-time":"2026-08-21T00:54:05Z","timestamp":1787273645000},"update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":3,"title":["An ab initio foundation model of wavefunctions that accurately describes chemical bond breaking"],"prefix":"10.1038","volume":"17","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-1624-7481","authenticated-orcid":false,"given":"Adam","family":"Foster","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Zeno","family":"Sch\u00e4tzle","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"P. Bern\u00e1t","family":"Szab\u00f3","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-7329-0585","authenticated-orcid":false,"given":"Lixue","family":"Cheng","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-7256-2892","authenticated-orcid":false,"given":"Jonas","family":"K\u00f6hler","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Gino","family":"Cassella","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-1487-2657","authenticated-orcid":false,"given":"Nicholas","family":"Gao","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jiawei","family":"Li","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-4169-9324","authenticated-orcid":false,"given":"Frank","family":"No\u00e9","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-2779-0749","authenticated-orcid":false,"given":"Jan","family":"Hermann","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"297","published-online":{"date-parts":[[2026,8,21]]},"reference":[{"key":"76604_CR1","doi-asserted-by":"publisher","first-page":"1049","DOI":"10.1103\/PhysRev.28.1049","volume":"28","author":"E Schr\u00f6dinger","year":"1926","unstructured":"Schr\u00f6dinger, E. An undulatory theory of the mechanics of atoms and molecules. Phys. Rev. 28, 1049 (1926).","journal-title":"Phys. Rev."},{"key":"76604_CR2","doi-asserted-by":"publisher","first-page":"5243","DOI":"10.1063\/1.475961","volume":"108","author":"SA Kucharski","year":"1998","unstructured":"Kucharski, S. A. & Bartlett, R. J.\u00a0Noniterative energy corrections through fifth-order to the coupled cluster singles and doubles method. J. Chem. Phys. 108, 5243\u20135254 (1998).","journal-title":"J. Chem. Phys."},{"key":"76604_CR3","doi-asserted-by":"publisher","first-page":"692","DOI":"10.1038\/s41570-023-00516-8","volume":"7","author":"J Hermann","year":"2023","unstructured":"Hermann, J. et al. Ab initio quantum chemistry with neural-network wavefunctions. Nat. Rev. Chem. 7, 692\u2013709 (2023).","journal-title":"Nat. Rev. Chem."},{"key":"76604_CR4","doi-asserted-by":"publisher","first-page":"891","DOI":"10.1038\/s41557-020-0544-y","volume":"12","author":"J Hermann","year":"2020","unstructured":"Hermann, J, Sch\u00e4tzle, Z. & Noe, F. Deep-neural-network solution of the electronic Schr\u00f6dinger equation. Nat. Chem. 12, 891\u2013897 (2020).","journal-title":"Nat. Chem."},{"key":"76604_CR5","doi-asserted-by":"publisher","first-page":"033429","DOI":"10.1103\/PhysRevResearch.2.033429","volume":"2","author":"D Pfau","year":"2020","unstructured":"Pfau, D., Spencer, J. S., Matthews, A. G. D. G. & Foulkes, W. M. C. Ab initio solution of the many-electron Schr\u00f6dinger equation with deep neural networks. Phys. Rev. Res. 2, 033429 (2020).","journal-title":"Phys. Rev. Res."},{"key":"76604_CR6","doi-asserted-by":"publisher","unstructured":"Spencer, J. S., Pfau, D., Botev, A. & Foulkes, W. M. C. Better, faster fermionic neural networks. Preprint at https:\/\/doi.org\/10.48550\/arXiv.2011.07125 (2020).","DOI":"10.48550\/arXiv.2011.07125"},{"key":"76604_CR7","unstructured":"von Glehn, I., Spencer, J. S. & Pfau, D. A self-attention ansatz for ab-initio quantum chemistry. In The Eleventh International Conference on Learning Representations. https:\/\/openreview.net\/forum?id=xveTeHVlF7j (ICLR, 2023)."},{"key":"76604_CR8","doi-asserted-by":"publisher","unstructured":"Gerard, L., Scherbela, M., Marquetand, P. & Grohs, P. Gold-standard solutions to the Schr\u00f6dinger equation using deep learning: How much physics do we need? In Advances in Neural Information Processing Systems (eds Oh, A. H., Agarwal, A., Belgrave, D. & Cho, K.) https:\/\/doi.org\/10.48550\/arXiv.2205.09438 (2022).","DOI":"10.48550\/arXiv.2205.09438"},{"key":"76604_CR9","doi-asserted-by":"publisher","first-page":"094108","DOI":"10.1063\/5.0157512","volume":"159","author":"Z Sch\u00e4tzle","year":"2023","unstructured":"Sch\u00e4tzle, Z., Szab\u00f3, P. B., Mezera, M., Hermann, J. & No\u00e9, F. DeepQMC: An open-source software suite for variational optimization of deep-learning molecular wave functions. J. Chem. Phys. 159, 094108 (2023).","journal-title":"J. Chem. Phys."},{"key":"76604_CR10","doi-asserted-by":"publisher","first-page":"111765","DOI":"10.1016\/j.jcp.2022.111765","volume":"474","author":"J Lin","year":"2023","unstructured":"Lin, J., Goldshlager, G. & Lin, L. Explicitly antisymmetrized neural network layers for variational Monte Carlo simulation. J. Comput. Phys. 474, 111765 (2023).","journal-title":"J. Comput. Phys."},{"key":"76604_CR11","unstructured":"Gao, N. & G\u00fcnnemann, S. Ab-initio potential energy surfaces by pairing gnns with neural wave functions. In International Conference on Learning Representations. https:\/\/openreview.net\/forum?id=apv504XsysP (ICLR, 2022)."},{"key":"76604_CR12","doi-asserted-by":"publisher","first-page":"013021","DOI":"10.1103\/PhysRevResearch.4.013021","volume":"4","author":"X Li","year":"2022","unstructured":"Li, X., Fan, C., Ren, W. & Chen, J. Fermionic neural network with effective core potential. Phys. Rev. Res. 4, 013021 (2022).","journal-title":"Phys. Rev. Res."},{"key":"76604_CR13","doi-asserted-by":"publisher","first-page":"235139","DOI":"10.1103\/PhysRevB.107.235139","volume":"107","author":"M Wilson","year":"2023","unstructured":"Wilson, M. et al. Neural network ansatz for periodic wave functions and the homogeneous electron gas. Phys. Rev. B 107, 235139 (2023).","journal-title":"Phys. Rev. B"},{"key":"76604_CR14","doi-asserted-by":"publisher","first-page":"036401","DOI":"10.1103\/PhysRevLett.130.036401","volume":"130","author":"G Cassella","year":"2023","unstructured":"Cassella, G. et al. Discovering quantum phase transitions with fermionic neural networks. Phys. Rev. Lett. 130, 036401 (2023).","journal-title":"Phys. Rev. Lett."},{"key":"76604_CR15","doi-asserted-by":"publisher","first-page":"148","DOI":"10.1038\/s42005-024-01613-w","volume":"7","author":"J Kim","year":"2024","unstructured":"Kim, J. et al. Neural-network quantum states for ultra-cold Fermi gases. Commun. Phys. 7, 148 (2024).","journal-title":"Commun. Phys."},{"key":"76604_CR16","doi-asserted-by":"publisher","first-page":"209","DOI":"10.1038\/s42256-024-00794-x","volume":"6","author":"R Li","year":"2024","unstructured":"Li, R. et al. A computational framework for neural network-based variational Monte Carlo with forward Laplacian. Nat. Mach. Intell. 6, 209\u2013219 (2024).","journal-title":"Nat. Mach. Intell."},{"key":"76604_CR17","first-page":"7922","volume":"20","author":"PB Szab\u00f3","year":"2024","unstructured":"Szab\u00f3, P. B., Sch\u00e4tzle, Z., Entwistle, M. T. & No\u00e9, F. An improved penalty-based excited-state variational monte carlo approach with deep-learning ansatzes. J. Chem. Theory Comput. 20, 7922\u20137935 (2024).","journal-title":"J. Chem. Theory Comput."},{"key":"76604_CR18","doi-asserted-by":"publisher","first-page":"eadn0137","DOI":"10.1126\/science.adn0137","volume":"385","author":"D Pfau","year":"2024","unstructured":"Pfau, D., Axelrod, s, Sutterud, H., von Glehn, I. & Spencer, J. S. Accurate computation of quantum excited states with neural networks. Science 385, eadn0137 (2024).","journal-title":"Science"},{"key":"76604_CR19","doi-asserted-by":"crossref","unstructured":"Cheng, L. et al. Highly accurate real-space electron densities with neural networks. J. Chem. Phys. 162, 034120 (2025).","DOI":"10.1063\/5.0236919"},{"key":"76604_CR20","unstructured":"Scherbela, M., Gerard, L., Gao, N., Grohs, P. & G\u00fcnnemann, S. Accurate Ab-initio neural-network solutions to large-scale electronic structure problems. Nat. Comput. Sci. https:\/\/ucrisportal.univie.ac.at\/en\/publications\/accurate-ab-initio-neural-network-solutions-to-large-scale-electr\/ (2025)."},{"key":"76604_CR21","doi-asserted-by":"publisher","first-page":"331","DOI":"10.1038\/s43588-022-00228-x","volume":"2","author":"M Scherbela","year":"2022","unstructured":"Scherbela, M., Reisenhofer, R., Gerard, L., Marquetand, P. & Grohs, P. Solving the electronic schr\u00f6dinger equation for multiple nuclear geometries with weight-sharing deep neural networks. Nat. Comput. Sci. 2, 331\u2013341 (2022).","journal-title":"Nat. Comput. Sci."},{"key":"76604_CR22","unstructured":"Gao, N. & G\u00fcnnemann, S. Generalizing neural wave functions. In Proc. International Conference on Machine Learning (2023)."},{"key":"76604_CR23","unstructured":"Gao, N. & G\u00fcnnemann, S. Neural Pfaffians: Solving many many-electron Schr\u00f6dinger equations. https:\/\/papers.nips.cc\/paper_files\/paper\/2024\/hash\/e2d28c3ec49945e1922943070ffd21e8-Abstract-Conference.html (NeurIPS, 2024)."},{"key":"76604_CR24","doi-asserted-by":"publisher","unstructured":"Sch\u00e4tzle, Z., Szab\u00f3, P. B., Cuzzocrea, A. & No\u00e9, F. Ab-initio simulation of excited-state potential energy surfaces with transferable deep quantum monte carlo. Nat. Comput. Sci. https:\/\/doi.org\/10.48550\/arXiv.2503.19847 (2025).","DOI":"10.48550\/arXiv.2503.19847"},{"key":"76604_CR25","doi-asserted-by":"crossref","unstructured":"Linteau, D., Moroni, S., Carleo, G. & Holzmann, M. Neural wave functions for high-pressure atomic hydrogen. Phy. Rev. Res. 8, L022007 (2026).","DOI":"10.1103\/t72h-tkcx"},{"key":"76604_CR26","doi-asserted-by":"crossref","unstructured":"Scherbela, M., Gerard, L. & Grohs, P. Variational monte carlo on a budget - fine-tuning pre-trained neural wavefunctions. In Advances in Neural Information Processing Systems, Vol 36 (2023).","DOI":"10.52202\/075280-1038"},{"key":"76604_CR27","doi-asserted-by":"publisher","DOI":"10.1038\/s41467-023-44216-9","volume":"15","author":"M Scherbela","year":"2024","unstructured":"Scherbela, M., Gerard, L. & Grohs, P. Towards a transferable fermionic neural wavefunction for molecules. Nat. Commun. 15, 120 (2024).","journal-title":"Nat. Commun."},{"key":"76604_CR28","doi-asserted-by":"crossref","unstructured":"Rende, R. et al. Foundation neural-networks quantum states as a unified Ansatz for multiple hamiltonians. Nat. Commun. 16, 7213 (2025).","DOI":"10.1038\/s41467-025-62098-x"},{"key":"76604_CR29","doi-asserted-by":"publisher","first-page":"98","DOI":"10.1002\/jlac.19284600106","volume":"460","author":"O Diels","year":"1928","unstructured":"Diels, O. & Alder, K. Synthesen in der hydroaromatischen reihe. Justus Liebigs Ann. der Chem. 460, 98\u2013122 (1928).","journal-title":"Justus Liebigs Ann. der Chem."},{"key":"76604_CR30","doi-asserted-by":"publisher","first-page":"652","DOI":"10.1002\/poc.3313","volume":"27","author":"C-X Cui","year":"2014","unstructured":"Cui, C.-X. & Liu, Y.-J. A thorough understanding of the Diels\u2013Alder reaction of 1,3-butadiene and ethylene. J. Phys. Org. Chem. 27, 652\u2013660 (2014).","journal-title":"J. Phys. Org. Chem."},{"key":"76604_CR31","doi-asserted-by":"publisher","first-page":"989","DOI":"10.1002\/qua.24933","volume":"115","author":"DLA Scarborough","year":"2015","unstructured":"Scarborough, D. L. A., Kobayashi, R., Thompson, C. D. & Izgorodina, E. I. Active space and basis set effects in CASPT2 models of the 1,3-butadiene-ethene cycloaddition and the 1,3-butadiene dimerization. Int. J. Quantum Chem. 115, 989\u20131001 (2015).","journal-title":"Int. J. Quantum Chem."},{"key":"76604_CR32","doi-asserted-by":"publisher","first-page":"1365","DOI":"10.1002\/cphc.200400104","volume":"5","author":"H Lischka","year":"2004","unstructured":"Lischka, H., Ventura, E. & Dallos, M. The Diels-Alder reaction of ethene and 1,3-butadiene: an extended multireference ab initio investigation. ChemPhysChem 5, 1365\u20131371 (2004).","journal-title":"ChemPhysChem"},{"key":"76604_CR33","doi-asserted-by":"publisher","first-page":"1878","DOI":"10.1021\/j100789a036","volume":"68","author":"M Uchiyama","year":"1964","unstructured":"Uchiyama, M., Tomioka, T. & Amano, A. Thermal decomposition of cyclohexene.  J. Phys. Chem. 68, 1878\u20131881 (1964).","journal-title":"J. Phys. Chem."},{"key":"76604_CR34","doi-asserted-by":"publisher","DOI":"10.1038\/s41597-021-01088-2","volume":"8","author":"VK Prasad","year":"2021","unstructured":"Prasad, V. K., Khalilian, M. H., Otero-de-la Roza, A. & DiLabio, G. A. Bse49, a diverse, high-quality benchmark dataset of separation energies of chemical bonds. Sci. Data 8, 300 (2021).","journal-title":"Sci. Data"},{"key":"76604_CR35","doi-asserted-by":"publisher","first-page":"151","DOI":"10.1002\/cpa.3160100201","volume":"10","author":"T Kato","year":"1957","unstructured":"Kato, T. On the eigenfunctions of many-particle systems in quantum mechanics. Commun. Pure Appl. Math. 10, 151\u2013177 (1957).","journal-title":"Commun. Pure Appl. Math."},{"key":"76604_CR36","doi-asserted-by":"publisher","unstructured":"Vaswani, A. et al. Attention is all you need. In 31st Conference on Neural Information Processing Systems https:\/\/doi.org\/10.48550\/arXiv.1706.03762 (NIPS, 2017).","DOI":"10.48550\/arXiv.1706.03762"},{"key":"76604_CR37","unstructured":"Paresh Kharya. TensorFloat-32 in the A100 GPU Accelerates AI Training, HPC up to 20x, https:\/\/blogs.nvidia.com\/blog\/tensorfloat-32-precision-format\/ (2020)."},{"key":"76604_CR38","unstructured":"Gao, N., K\u00f6hler, J., Foster, A. & Szab\u00f3, P. B. folx - Forward Laplacian for JAX, https:\/\/github.com\/microsoft\/folx (2023)."},{"key":"76604_CR39","doi-asserted-by":"publisher","unstructured":"Dao, T., Fu, D., Ermon, S., Rudra, A. & R\u00e9, C. FlashAttention: Fast and memory-efficient exact attention with IO-awareness. Preprint at https:\/\/doi.org\/10.48550\/arXiv.2205.14135 (2022).","DOI":"10.48550\/arXiv.2205.14135"},{"key":"76604_CR40","unstructured":"The JAX Team. Pallas: a JAX kernel language, https:\/\/docs.jax.dev\/en\/latest\/pallas\/index.html (2023)."},{"key":"76604_CR41","doi-asserted-by":"crossref","unstructured":"Ulf Grenander and Michael I Miller. Representations of knowledge in complex systems. J. R. Stat. Soc. Ser. B Stat. Methodol. 56, 549\u2013581 (1994).","DOI":"10.1111\/j.2517-6161.1994.tb02000.x"},{"key":"76604_CR42","doi-asserted-by":"publisher","first-page":"460","DOI":"10.1021\/ct2006852","volume":"8","author":"W Jiang","year":"2012","unstructured":"Jiang, W., DeYonker, N. J. & Wilson, A. K. Multireference character for 3d transition-metal-containing molecules. J. Chem. Theory Comput. 8, 460\u2013468 (2012).","journal-title":"J. Chem. Theory Comput."},{"key":"76604_CR43","doi-asserted-by":"crossref","unstructured":"Bengio, Y., Louradour, J., Collobert, R. & Weston, J. Curriculum learning. In Proceedings of the 26th Annual International Conference on Machine Learning, pages 41\u201348, 2009.","DOI":"10.1145\/1553374.1553380"},{"key":"76604_CR44","doi-asserted-by":"publisher","unstructured":"Foster, A. et al. oneQMC Repository, https:\/\/doi.org\/10.5281\/zenodo.21160150 (2026).","DOI":"10.5281\/zenodo.21160150"},{"key":"76604_CR45","doi-asserted-by":"publisher","first-page":"198","DOI":"10.1039\/df9511000198","volume":"10","author":"D Rowley","year":"1951","unstructured":"Rowley, D. & Steiner, H. Kinetics of diene reactions at high temperatures. Discuss. Faraday Soc. 10, 198\u2013213 (1951).","journal-title":"Discuss. Faraday Soc."}],"container-title":["Nature Communications"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.nature.com\/articles\/s41467-026-76604-2","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/www.nature.com\/articles\/s41467-026-76604-2.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/www.nature.com\/articles\/s41467-026-76604-2.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,21]],"date-time":"2026-09-21T11:01:49Z","timestamp":1789988509000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.nature.com\/articles\/s41467-026-76604-2"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,8,21]]},"references-count":45,"journal-issue":{"issue":"1","published-online":{"date-parts":[[2026,12]]}},"alternative-id":["76604"],"URL":"https:\/\/doi.org\/10.1038\/s41467-026-76604-2","relation":{},"ISSN":["2041-1723"],"issn-type":[{"value":"2041-1723","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,8,21]]},"assertion":[{"value":"2 October 2025","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"31 July 2026","order":2,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"21 August 2026","order":3,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"The authors declare no competing interests.","order":1,"name":"Ethics","label":"Competing interests","group":{"name":"EthicsHeading","label":"Ethics"}}],"article-number":"9976"}}