TabPFN-3: Technical Report

@misc{grinsztajn2026tabpfn3,
  title = {{TabPFN-3: Technical Report}},
  author = {Léo Grinsztajn and Klemens Flöge and Oscar Key and Felix Birkel and Philipp Jund and Brendan Roof and Mihir Manium and Shi Bin Hoo and Magnus Bühler and Anurag Garg and Dominik Safaric and Jake Robertson and Benjamin Jäger and Simone Alessi and Adrian Hayler and Vladyslav Moroshan and Lennart Purucker and Philipp Singer and Alan Arazi and Julien Siems and Jan Hendrik Metzen and Georg Grab and Nick Erickson and Siyuan Guo and Eliott Kalfon and Simon Bing and David Salinas and Clara Cornu and Lilly Charlotte Wehrhahn and Diana Kriuchkova and Kursat Kaya and Lydia Sidhoum and Marie Salmon and Jerry Chen and Madelon Hulsebos and Yann LeCun and Samuel Müller and Bernhard Schölkopf and Sauraj Gambhir and Noah Hollmann and Frank Hutter},
  howpublished = {arXiv},
  year = {2026},
  month = nov,
  doi = {10.48550/arxiv.2605.13986},
  eprint = {2605.13986},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2605.13986},
}

TY  - UNPB
TI  - TabPFN-3: Technical Report
AU  - Léo Grinsztajn
AU  - Klemens Flöge
AU  - Oscar Key
AU  - Felix Birkel
AU  - Philipp Jund
AU  - Brendan Roof
AU  - Mihir Manium
AU  - Shi Bin Hoo
AU  - Magnus Bühler
AU  - Anurag Garg
AU  - Dominik Safaric
AU  - Jake Robertson
AU  - Benjamin Jäger
AU  - Simone Alessi
AU  - Adrian Hayler
AU  - Vladyslav Moroshan
AU  - Lennart Purucker
AU  - Philipp Singer
AU  - Alan Arazi
AU  - Julien Siems
AU  - Jan Hendrik Metzen
AU  - Georg Grab
AU  - Nick Erickson
AU  - Siyuan Guo
AU  - Eliott Kalfon
AU  - Simon Bing
AU  - David Salinas
AU  - Clara Cornu
AU  - Lilly Charlotte Wehrhahn
AU  - Diana Kriuchkova
AU  - Kursat Kaya
AU  - Lydia Sidhoum
AU  - Marie Salmon
AU  - Jerry Chen
AU  - Madelon Hulsebos
AU  - Yann LeCun
AU  - Samuel Müller
AU  - Bernhard Schölkopf
AU  - Sauraj Gambhir
AU  - Noah Hollmann
AU  - Frank Hutter
PY  - 2026
DA  - 2026/11/03
PB  - arXiv
DO  - 10.48550/arxiv.2605.13986
UR  - https://arxiv.org/abs/2605.13986
AB  - 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. On more diverse datasets, TabPFN-3 ranks first on datasets with many classes, and beats 8-hour-tuned gradient-boosted-tree baselines on datasets up to 1M training rows and 200 features. 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 extends the capabilities of our models, enabling SOTA prediction on relational data (new SOTA foundation model on RelBenchV1) and tabular-text data (SOTA on TabSTAR via TabPFN-3-Plus); and improves existing integrations: a specialized checkpoint, TabPFN-TS-3, ranks 2nd on the time-series benchmark fev-bench, and SHAP-value computation is up to 120x faster. TabPFN-3 achieves this performance while being up to 20x faster than TabPFN-2.5. In addition, a reduced KV cache and row-chunking scale to 1M rows on one H100 with fast inference speed.
ER  -