# tabular foundation — RSS Amplifier

Recent posts from the 2 feeds in the RSS Amplifier directory that cover tabular foundation.

Page: <https://rssamplifier.com/topics/tabular-foundation>  
Feed: <https://rssamplifier.com/topics/tabular-foundation.md>

---

## [Nori-30M: A Small Open Model That Beats Tuned CatBoost at Regression](https://aihorizonforecast.substack.com/p/nori-30m-a-small-open-model-that)

_2026-08-10 · Nikos Kafritsas · AI Horizon Forecast_

A head-to-head regression benchmark of Nori-30M, TabPFN-3, TabICL-v2, and TabFM against tuned and untuned CatBoost.

## [Do Tabular Foundation Models Actually Beat CatBoost? - Part 2](https://aihorizonforecast.substack.com/p/do-tabular-foundation-models-actually)

_2026-08-07 · Nikos Kafritsas · AI Horizon Forecast_

A head-to-head benchmark of TabPFN-3, TabICL-v2, and TabFM against tuned and untuned CatBoost.

## [Tabular Foundation Models Are Not Plug-and-Play - Part 1](https://aihorizonforecast.substack.com/p/tabular-foundation-models-are-not)

_2026-08-02 · Nikos Kafritsas · AI Horizon Forecast_

The pretrained models beating gradient boosting come with hidden settings. A guide to TabPFN-3, TabICL-v2, and TabFM

## [Trends in tabular foundation research](https://mindfulmodeler.substack.com/p/trends-in-tabular-foundation-research)

_2026-07-22 · Christoph Molnar · Mindful Modeler_

Based on 151 papers from ICML workshop "Foundation Models for Structured Data"

## [Time for a change](https://mindfulmodeler.substack.com/p/time-for-a-change)

_2026-07-16 · Christoph Molnar · Mindful Modeler_

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)

## [Toto-2.0: 5 Silent Failure Modes to Avoid for Full Performance (+Tutorial)](https://aihorizonforecast.substack.com/p/toto-20-5-silent-failure-modes-to)

_2026-07-10 · Nikos Kafritsas · AI Horizon Forecast_

Toto-2.0 is a top-tier foundation model - here is how to actually leverage its full potential

## [Toto-2.0: Time Series Forecasting Finally Scales Like LLMs](https://aihorizonforecast.substack.com/p/toto-20-time-series-forecasting-finally)

_2026-07-08 · Nikos Kafritsas · AI Horizon Forecast_

5 model variants, one training recipe, clean scaling from 4M to 2.5B

## [TabFM minus the hype](https://mindfulmodeler.substack.com/p/tabfm-minus-the-hype)

_2026-07-07 · Christoph Molnar · Mindful Modeler_

top performance on TabArena; large model; slow inference; non-commercial license

## [Which ML models produce the best quantile estimates?](https://mindfulmodeler.substack.com/p/which-regressor-produces-the-best)

_2026-06-30 · Christoph Molnar · Mindful Modeler_

Results from ScoringBench

## [Toto-2.0: Learning to Forecast Sparse Time Series Without Training](https://aihorizonforecast.substack.com/p/toto-20-learning-to-forecast-sparse)

_2026-06-25 · Nikos Kafritsas · AI Horizon Forecast_

Understanding, benchmarking, and using a foundation model for sparse and intermittent time series forecasting without training

## [CLI-first decentralized GPU compute (Sponsored)](https://crawlproof.com/a/qkMOGAxqyQtY)

_2026-06-25 · **Sponsored**_

Pay workers or run your GPU for FFmpeg transcode and AI inference.

## [When trees still beat tabular foundation models](https://mindfulmodeler.substack.com/p/when-trees-still-beat-tabular-foundation)

_2026-06-23 · Christoph Molnar · Mindful Modeler_

As you might have noticed, I’m rather optimistic about tabular foundation models.

## [What is TabPFN's Thinking mode?](https://mindfulmodeler.substack.com/p/what-is-tabpfns-thinking-mode)

_2026-06-16 · Christoph Molnar · Mindful Modeler_

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 datasets.

## [How TabICL and TabPFN handle missing values](https://mindfulmodeler.substack.com/p/how-tabicl-and-tabpfn-handle-missing)

_2026-06-09 · Christoph Molnar · Mindful Modeler_

You can give training or test data with missing values to the tabular foundation models TabPFN and TabICL, and the prediction will “just work”.

## [Toto-2.0: The Upgraded Time-Series Foundation Model (+Retail Forecasting Tutorial)](https://aihorizonforecast.substack.com/p/toto-20-the-upgraded-time-series)

_2026-06-07 · Nikos Kafritsas · AI Horizon Forecast_

Datadog's new open-weight SOTA model tested on a retail forecasting project.

## [TabPFN and TabICL are not everything](https://mindfulmodeler.substack.com/p/tabpfn-and-tabicl-are-not-everything)

_2026-06-02 · Christoph Molnar · Mindful Modeler_

An overview of other families for tabular foundation models

## [I’m writing a book on Tabular Foundation Models](https://mindfulmodeler.substack.com/p/im-writing-a-book-on-tabular-foundation)

_2026-05-26 · Christoph Molnar · Mindful Modeler_

tl;dr: In-progress book here: tabularfoundationmodels.com

## [Tabular ML is entering a new benchmark era](https://mindfulmodeler.substack.com/p/tabular-ml-is-entering-a-new-benchmark)

_2026-05-19 · Christoph Molnar · Mindful Modeler_

From static and narrow benchmarks to live, capability-driven evaluation

## [Time Series Foundation Models: A Deep Dive into Strengths and Limitations](https://aihorizonforecast.substack.com/p/time-series-foundation-models-a-deep)

_2026-04-24 · AI Horizon Forecast_

What works, what doesn’t, and how to make them work beyond the hype

## [TabICLV2 for Time Series Forecasting: A Complete Tutorial](https://aihorizonforecast.substack.com/p/tabiclv2-for-time-series-forecasting)

_2026-04-13 · Nikos Kafritsas · AI Horizon Forecast_

Turning a leading tabular foundation model into a zero-shot forecaster.

## [How to Build an AI Document Assistant App with Docling and LangChain](https://aihorizonforecast.substack.com/p/how-to-build-an-ai-document-assistant)

_2026-02-19 · Nikos Kafritsas · AI Horizon Forecast_

Part 3 on integrating a chatbot with your personal document collection

## [Point once, get highlights (Sponsored)](https://crawlproof.com/a/k3XThGYh3yQO)

_2026-02-19 · **Sponsored**_

Get minutes worth watching from a whole game — point once

## [Part 2: Making Documents Searchable with Vector Stores and Retrieval](https://aihorizonforecast.substack.com/p/part-2-making-documents-searchable)

_2026-02-09 · Nikos Kafritsas · AI Horizon Forecast_

Building RAG pipelines with Chroma and LangChain

## [Docling: Extracting Structured Data from PDFs and Documents](https://aihorizonforecast.substack.com/p/docling-extracting-structured-data)

_2026-02-03 · Nikos Kafritsas · AI Horizon Forecast_

Part 1 on integrating a chatbot with your personal document collection

## [MOIRAI-2 Technical Report: Discussing the Recent Advances in Foundation Time Series Models](https://aihorizonforecast.substack.com/p/moirai-2-technical-report-discussing)

_2025-11-29 · Nikos Kafritsas · AI Horizon Forecast_

Exploring the upgraded foundation time series model.

## [Chronos-2: Cold-Start Forecasting with Short Histories and No Training — A Practical Tutorial](https://aihorizonforecast.substack.com/p/chronos-2-cold-start-forecasting)

_2025-11-16 · Nikos Kafritsas · AI Horizon Forecast_

An interesting case with the Chronos-2 foundation model

## [Chronos-2: Multivariate Zero-Shot Forecasting with External Covariates](https://aihorizonforecast.substack.com/p/chronos-2-multivariate-zero-shot)

_2025-11-14 · Nikos Kafritsas · AI Horizon Forecast_

Amazon Research's upgraded model smashed the benchmarks

## [Chronos-2: Zeroshot Multivariate Forecasting with Covariates — A Practical Tutorial for Retail Forecasting](https://aihorizonforecast.substack.com/p/chronos-2-zeroshot-multivariate-forecasting)

_2025-11-06 · Nikos Kafritsas · AI Horizon Forecast_

The popular foundation forecasting model received an upgrade!

## [TimesFM-2.5 + External Covariates: A Practical Retail Forecasting Tutorial](https://aihorizonforecast.substack.com/p/timesfm-25-external-covariates-a)

_2025-10-30 · Nikos Kafritsas · AI Horizon Forecast_

Enhancing the popular foundation model with external covariates

## [Transformers, Time Series, and the Myth of Permutation Invariance](https://aihorizonforecast.substack.com/p/transformers-time-series-and-the)

_2025-10-17 · Nikos Kafritsas · AI Horizon Forecast_

Beware the false prophets of forecasting!

## [Adapting Time-Series Foundation Models at Inference Time](https://aihorizonforecast.substack.com/p/adapting-time-series-foundation-models)

_2025-10-11 · Nikos Kafritsas · AI Horizon Forecast_

TimesFM-ICF: Letting foundation models learn as they predict!

## [Find what they just said (Sponsored)](https://crawlproof.com/a/CsFpZf3i1FfY)

_2025-10-11 · **Sponsored**_

Surface recent public signals so you can send policy-aware, relevant outreach

## [TimesFM-2.5: Hands-On Tutorial with Google's Upgraded Forecasting Model](https://aihorizonforecast.substack.com/p/timesfm-25-hands-on-tutorial-with)

_2025-10-04 · Nikos Kafritsas · AI Horizon Forecast_

Google's foundation forecasting model received its 3rd upgrade!

