{"ID":2879267,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2508.16067","arxiv_id":"2508.16067","title":"Training a Foundation Model for Materials on a Budget","abstract":"Foundation models for materials modeling are advancing quickly, but their training remains expensive, often placing state-of-the-art methods out of reach for many research groups. We introduce Nequix, a compact E(3)-equivariant potential that pairs a simplified NequIP design with modern training practices, including equivariant root-mean-square layer normalization and the Muon optimizer, to retain accuracy while substantially reducing compute requirements. Nequix has 700K parameters and was trained in 100 A100 GPU-hours. On the Matbench-Discovery and MDR Phonon benchmarks, Nequix ranks third overall while requiring a 20 times lower training cost than most other methods, and it delivers two orders of magnitude faster inference speed than the current top-ranked model. We release model weights and fully reproducible codebase at https://github.com/atomicarchitects/nequix.","short_abstract":"Foundation models for materials modeling are advancing quickly, but their training remains expensive, often placing state-of-the-art methods out of reach for many research groups. We introduce Nequix, a compact E(3)-equivariant potential that pairs a simplified NequIP design with modern training practices, including eq...","url_abs":"https://arxiv.org/abs/2508.16067","url_pdf":"https://arxiv.org/pdf/2508.16067v2","authors":"[\"Teddy Koker\",\"Mit Kotak\",\"Tess Smidt\"]","published":"2025-08-22T03:38:06Z","proceeding":"physics.comp-ph","tasks":"[\"physics.comp-ph\",\"cs.LG\"]","methods":"[]","has_code":false,"code_links":[{"ID":610570,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_id":2879267,"paper_url":"https://arxiv.org/abs/2508.16067","paper_title":"Training a Foundation Model for Materials on a Budget","repo_url":"https://github.com/atomicarchitects/nequix","is_official":false,"mentioned_in_paper":false,"mentioned_in_github":true,"github_stars":0}]}