TabPFN-3: Technical Report

Léo Grinsztajn, Klemens Flöge, Oscar Key, Felix Birkel, Philipp Jund, Brendan Roof, Mihir Manium, Shi Bin Hoo, Magnus Bühler, Anurag Garg, Dominik Safaric, Jake Robertson, Benjamin Jäger, Simone Alessi, Adrian Hayler, Vladyslav Moroshan, Lennart Purucker, Philipp Singer, Alan Arazi, Julien Siems, Jan Hendrik Metzen, Georg Grab, Nick Erickson, Siyuan Guo, Eliott Kalfon, Simon Bing, David Salinas, Clara Cornu, Lilly Charlotte Wehrhahn, Diana Kriuchkova, Kursat Kaya, Lydia Sidhoum, Marie Salmon, Jerry Chen, Madelon Hulsebos, Yann LeCun, Samuel Müller, Bernhard Schölkopf, Sauraj Gambhir, Noah Hollmann, Frank Hutter

arXiv preprint·May 2026preprint

Abstract

Tabular data underpins most high-value prediction problems in science and industry, and TabPFN has driven the foundation model revolution for this modality. Designed with feedback from our users, TabPFN-3 builds on this foundation to scale state-of-the-art performance to datasets with 1M training rows and substantially reduce training and inference time. Pretrained exclusively on synthetic data from our prior, TabPFN-3 dramatically pushes the frontier of tabular prediction and brings substantial gains on time series, relational, and tabular-text data. On the standard tabular benchmark TabArena, a forward pass of TabPFN-3 outperforms all other models, including tuned and ensembled baselines, by a significant margin, and pareto-dominates the speed/performance frontier. TabPFN-3 introduces test-time compute scaling to tabular foundation models. Our API offering TabPFN-3-Plus (Thinking) exploits this to beat all non-TabPFN models by over 200 Elo on TabArena, rising to 420 Elo on the largest data subset, and outperforms AutoGluon 1.5 extreme while being 10x faster, without using LLMs, real data, internet search or any other model besides TabPFN. TabPFN-3 achieves this performance while being up to 20x faster than TabPFN-2.5.

TabPFN-3 is the clearest expression of the idea that drew me to Prior Labs: rather than writing down a prior over a network’s parameters, you specify it in function space by describing what tabular data ought to look like, sample datasets from that prior, and train a model to do the inference in a single forward pass.

The practical consequences are large. TabPFN-3 scales to a million training rows, runs up to 20x faster than TabPFN-2.5, and a single forward pass outperforms tuned and ensembled baselines on TabArena. It also brings test-time compute scaling to tabular models, and extends beyond flat tables to relational and tabular-text data.