[![Mindful Modeler](https://substackcdn.com/image/fetch/$s_!WtQm!,w_40,h_40,c_fill,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2Fc3dc1d90-fb9f-4e3c-80d5-756a5d6e8495_500x500.png)](https://mindfulmodeler.substack.com/)
# [Mindful Modeler](https://mindfulmodeler.substack.com/)
SubscribeSign in
[Home](https://mindfulmodeler.substack.com/)
[Tabular Foundation Models](https://mindfulmodeler.substack.com/t/tabular-foundation-models)
[Archive](https://mindfulmodeler.substack.com/archive)
[About](https://mindfulmodeler.substack.com/about)
## Page not found
LatestTopDiscussions
[Trends in tabular foundation research](https://mindfulmodeler.substack.com/p/trends-in-tabular-foundation-research)
[Based on 151 papers from ICML workshop "Foundation Models for Structured Data"](https://mindfulmodeler.substack.com/p/trends-in-tabular-foundation-research)
Jul 22 • [Christoph Molnar](https://substack.com/@christophmolnar)
28
3
[Time for a change](https://mindfulmodeler.substack.com/p/time-for-a-change)
[After 4 years of writing, I'm exploring what's next for me in ML / data science / AI (or whatever you want to call it)](https://mindfulmodeler.substack.com/p/time-for-a-change)
Jul 16 • [Christoph Molnar](https://substack.com/@christophmolnar)
32
8
![](../images/013.jpeg)
[TabFM minus the hype](https://mindfulmodeler.substack.com/p/tabfm-minus-the-hype)
[top performance on TabArena; large model; slow inference; non-commercial license](https://mindfulmodeler.substack.com/p/tabfm-minus-the-hype)
Jul 7 • [Christoph Molnar](https://substack.com/@christophmolnar)
34
### June 2026
[Which ML models produce the best quantile estimates?](https://mindfulmodeler.substack.com/p/which-regressor-produces-the-best)
[Results from ScoringBench](https://mindfulmodeler.substack.com/p/which-regressor-produces-the-best)
Jun 30 • [Christoph Molnar](https://substack.com/@christophmolnar)
14
2
![](../images/body-9a9aa4446c242a528920.png)
[When trees still beat tabular foundation models](https://mindfulmodeler.substack.com/p/when-trees-still-beat-tabular-foundation)
[As you might have noticed, I’m rather optimistic about tabular foundation models.](https://mindfulmodeler.substack.com/p/when-trees-still-beat-tabular-foundation)
Jun 23 • [Christoph Molnar](https://substack.com/@christophmolnar)
30
8
1
![](../images/body-046df6a26d0cb93d2488.png)
[What is TabPFN's Thinking mode?](https://mindfulmodeler.substack.com/p/what-is-tabpfns-thinking-mode)
[When I read the TabPFN-3 Technical Report, the benchmark for “TabPFN-3-Thinking” stood out: it appeared at the top of the TabArena benchmark for large…](https://mindfulmodeler.substack.com/p/what-is-tabpfns-thinking-mode)
Jun 16 • [Christoph Molnar](https://substack.com/@christophmolnar)
16
1
2
![](../images/body-ee0673943cbdb01b8702.png)
[How TabICL and TabPFN handle missing values](https://mindfulmodeler.substack.com/p/how-tabicl-and-tabpfn-handle-missing)
[You can give training or test data with missing values to the tabular foundation models TabPFN and TabICL, and the prediction will “just work”.](https://mindfulmodeler.substack.com/p/how-tabicl-and-tabpfn-handle-missing)
Jun 9 • [Christoph Molnar](https://substack.com/@christophmolnar)
23
1
3
[TabPFN and TabICL are not everything](https://mindfulmodeler.substack.com/p/tabpfn-and-tabicl-are-not-everything)
[An overview of other families for tabular foundation models](https://mindfulmodeler.substack.com/p/tabpfn-and-tabicl-are-not-everything)
Jun 2 • [Christoph Molnar](https://substack.com/@christophmolnar)
22
2
![](../images/body-2f890e121a5085a082b6.png)
### May 2026
[I’m writing a book on Tabular Foundation Models](https://mindfulmodeler.substack.com/p/im-writing-a-book-on-tabular-foundation)
[tl;dr: In-progress book here: tabularfoundationmodels.com](https://mindfulmodeler.substack.com/p/im-writing-a-book-on-tabular-foundation)
May 26 • [Christoph Molnar](https://substack.com/@christophmolnar)
33
12
2
![](../images/body-12982a727befec6b17c8.png)
[Tabular ML is entering a new benchmark era](https://mindfulmodeler.substack.com/p/tabular-ml-is-entering-a-new-benchmark)
[From static and narrow benchmarks to live, capability-driven evaluation](https://mindfulmodeler.substack.com/p/tabular-ml-is-entering-a-new-benchmark)
May 19 • [Christoph Molnar](https://substack.com/@christophmolnar)
16
[How to make Tabular Foundation Model inference faster](https://mindfulmodeler.substack.com/p/making-tabular-foundation-models)
[GPU, KV caching, context optimization, and other inference tricks](https://mindfulmodeler.substack.com/p/making-tabular-foundation-models)
May 12 • [Christoph Molnar](https://substack.com/@christophmolnar)
21
2
1
![](../images/body-481fb13c4b2e21ea704e.png)
[Time series forecasting with tabular foundation models](https://mindfulmodeler.substack.com/p/time-series-forecasting-with-tabular)
[This post is a quick primer on using tabular foundation models for time series forecasting.](https://mindfulmodeler.substack.com/p/time-series-forecasting-with-tabular)
May 5 • [Christoph Molnar](https://substack.com/@christophmolnar)
19
4
3
![](../images/body-448d720f9092778758f7.png)
### April 2026
[Train or Test for Feature Effect Estimation? We Finally Have an Answer](https://mindfulmodeler.substack.com/p/train-or-test-for-feature-effect)
[A guest post by XAI researcher Timo Heiß](https://mindfulmodeler.substack.com/p/train-or-test-for-feature-effect)
Apr 28 • [Christoph Molnar](https://substack.com/@christophmolnar)
16
3
3
![](../images/body-f888e7ab5c600aa1582b.png)
[Context is the new training](https://mindfulmodeler.substack.com/p/context-is-the-new-training)
[Tabular foundation models replace retraining with editable inference-time data](https://mindfulmodeler.substack.com/p/context-is-the-new-training)
Apr 21 • [Christoph Molnar](https://substack.com/@christophmolnar)
39
1
3
![](../images/body-2a3aee0dac04ef1123b5.png)
[Regression should predict full distributions](https://mindfulmodeler.substack.com/p/regression-should-predict-full-distributions)
[A "hidden" feature of tabular foundation models](https://mindfulmodeler.substack.com/p/regression-should-predict-full-distributions)
Apr 14 • [Christoph Molnar](https://substack.com/@christophmolnar)
39
2
1
![](../images/body-60cc0828977045205fcc.png)
### March 2026
[TabArena explained](https://mindfulmodeler.substack.com/p/tabarena-explained)
[What chess ratings tell us about ML models](https://mindfulmodeler.substack.com/p/tabarena-explained)
Mar 31 • [Christoph Molnar](https://substack.com/@christophmolnar)
20
4
1
![](../images/body-416bec0248c6f5d3e732.png)
[The interpretability tax on tabular foundation models](https://mindfulmodeler.substack.com/p/tabular-foundation-models-break-the)
[Model-agnostic interpretability for tabular foundation models](https://mindfulmodeler.substack.com/p/tabular-foundation-models-break-the)
Mar 24 • [Christoph Molnar](https://substack.com/@christophmolnar)
25
9
1
![](../images/body-66941eb114875939ad1f.png)
[I’m betting on tabular foundation models](https://mindfulmodeler.substack.com/p/im-betting-on-tabular-foundation)
[The shipping container moment of data science](https://mindfulmodeler.substack.com/p/im-betting-on-tabular-foundation)
Mar 17 • [Christoph Molnar](https://substack.com/@christophmolnar)
66
9
9
![](../images/body-3625aa898ddc4395ee89.png)
[The Random Forest of the 2030s?](https://mindfulmodeler.substack.com/p/the-random-forest-of-the-2030s)
[Three scenarios for the future of tabular ML, plus one that makes them all irrelevant](https://mindfulmodeler.substack.com/p/the-random-forest-of-the-2030s)
Mar 3 • [Christoph Molnar](https://substack.com/@christophmolnar)
24
2
### February 2026
[A regression example with TabICLv2](https://mindfulmodeler.substack.com/p/a-regression-example-with-tabiclv2)
[Strong performance at a steep interpretability cost (at least on CPU)](https://mindfulmodeler.substack.com/p/a-regression-example-with-tabiclv2)
Feb 24 • [Christoph Molnar](https://substack.com/@christophmolnar)
27
3
1
![](../images/body-71a8c6582e2d2490ec96.png)
[The state of Tabular Foundation Models (2026)](https://mindfulmodeler.substack.com/p/the-state-of-tabular-foundation-models)
[Pre-trained models, startups, and recommendation.](https://mindfulmodeler.substack.com/p/the-state-of-tabular-foundation-models)
Feb 17 • [Christoph Molnar](https://substack.com/@christophmolnar)
41
7
4
![](../images/body-a7463d6bf7a12481dfa3.png)
[How Tabular Foundation Models learn without real data](https://mindfulmodeler.substack.com/p/how-tabular-foundation-models-are)
[The pre-training recipe behind TFMs](https://mindfulmodeler.substack.com/p/how-tabular-foundation-models-are)
Feb 10 • [Christoph Molnar](https://substack.com/@christophmolnar)
36
2
4
![](../images/body-bc538b4e7ffa12b459cf.png)
[The architecture behind TabPFN](https://mindfulmodeler.substack.com/p/the-architecture-behind-tabpfn)
[How transformers learn from entire tables](https://mindfulmodeler.substack.com/p/the-architecture-behind-tabpfn)
Feb 3 • [Christoph Molnar](https://substack.com/@christophmolnar)
43
7
1
![](../images/body-b47de25039bd1c8bf330.png)
### January 2026
[How PFNs make tabular foundation models work](https://mindfulmodeler.substack.com/p/how-pfns-make-tabular-foundation)
[Bayesian logic, synthetic priors, and in-context learning](https://mindfulmodeler.substack.com/p/how-pfns-make-tabular-foundation)
Jan 21 • [Christoph Molnar](https://substack.com/@christophmolnar)
62
2
4
![](../images/body-da05da41155b04c02fdb.png)
© 2026 Christoph Molnar · [Privacy](https://substack.com/privacy) ∙ [Terms](https://substack.com/tos) ∙ [Collection notice](https://substack.com/ccpa#personal-data-collected)
[ Start your Substack](https://substack.com/signup?utm_source=substack&utm_medium=web&utm_content=footer)[Get the app](https://substack.com/app/app-store-redirect?utm_campaign=app-marketing&utm_content=web-footer-button)
[Substack](https://substack.com) is the home for great culture