{"id":14275,"date":"2026-09-07T03:43:35","date_gmt":"2026-09-07T07:43:35","guid":{"rendered":"https:\/\/www.predictiveanalyticsworld.com\/machinelearningtimes\/?p=14275"},"modified":"2026-09-07T03:43:35","modified_gmt":"2026-09-07T07:43:35","slug":"tabular-llms-an-introduction-to-the-foundation-models-that-predict-your-spreadsheet","status":"publish","type":"post","link":"https:\/\/www.predictiveanalyticsworld.com\/machinelearningtimes\/tabular-llms-an-introduction-to-the-foundation-models-that-predict-your-spreadsheet\/14275\/","title":{"rendered":"Tabular LLMs: An Introduction to the Foundation Models That Predict Your Spreadsheet"},"content":{"rendered":"Originally published on\u00a0towards data science, July 24, 2026. A 28M-parameter model you don\u2019t train beats tuned XGBoost. An introduction to tabular foundation models, with an independent reproduction. A tabular foundation model predicts the missing column of any table, zero-shot, the way a language model completes text. On the main community benchmark, every single-model entry above the best tuned gradient-boosted tree is now one of these. This post explains what they are, verifies the strongest one with unrestricted open weights on my own hardware, and maps where the trees still win. There is a fairly new class of model <a href=\"https:\/\/www.predictiveanalyticsworld.com\/machinelearningtimes\/tabular-llms-an-introduction-to-the-foundation-models-that-predict-your-spreadsheet\/14275\/\" class=\"more-link\">(more&hellip;)<\/a>","protected":false},"excerpt":{"rendered":"<p>Originally published on\u00a0towards data science, July 24, 2026. A 28M-parameter model you don\u2019t train beats tuned XGBoost. An introduction to tabular foundation models, with an independent reproduction. A tabular foundation model predicts the missing column of any table, zero-shot, the way a language model completes text. On the main community benchmark, every single-model entry above [&hellip;]<\/p>\n","protected":false},"author":13468,"featured_media":0,"comment_status":"closed","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":"","_links_to":"","_links_to_target":""},"categories":[11,48],"tags":[],"class_list":["post-14275","post","type-post","status-publish","format-standard","hentry","category-industry-news","category-left-hand"],"_links":{"self":[{"href":"https:\/\/www.predictiveanalyticsworld.com\/machinelearningtimes\/wp-json\/wp\/v2\/posts\/14275","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.predictiveanalyticsworld.com\/machinelearningtimes\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.predictiveanalyticsworld.com\/machinelearningtimes\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.predictiveanalyticsworld.com\/machinelearningtimes\/wp-json\/wp\/v2\/users\/13468"}],"replies":[{"embeddable":true,"href":"https:\/\/www.predictiveanalyticsworld.com\/machinelearningtimes\/wp-json\/wp\/v2\/comments?post=14275"}],"version-history":[{"count":1,"href":"https:\/\/www.predictiveanalyticsworld.com\/machinelearningtimes\/wp-json\/wp\/v2\/posts\/14275\/revisions"}],"predecessor-version":[{"id":14276,"href":"https:\/\/www.predictiveanalyticsworld.com\/machinelearningtimes\/wp-json\/wp\/v2\/posts\/14275\/revisions\/14276"}],"wp:attachment":[{"href":"https:\/\/www.predictiveanalyticsworld.com\/machinelearningtimes\/wp-json\/wp\/v2\/media?parent=14275"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.predictiveanalyticsworld.com\/machinelearningtimes\/wp-json\/wp\/v2\/categories?post=14275"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.predictiveanalyticsworld.com\/machinelearningtimes\/wp-json\/wp\/v2\/tags?post=14275"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}