Sakhanda Wire
NVDA $153.32 -1.81% MSFT $505.84 -2.35% GOOGL $280.31 -1.20% META $621.44 -0.79% AMZN $257.31 -0.17%
← Back to the news

NVIDIA Releases Kumo Tabular Foundation Models for Table Prediction

NVIDIA has released Kumo Tabular, a family of tabular foundation models for classification and regression. The models use labeled rows as context and predict previously unseen rows in a single forward pass, without task-specific training, hyperparameter tuning, or feature engineering. Small, Medium, and Large versions range from about 28M to 215M parameters.

Kumo Tabular is available through NVIDIA’s open-source structured-data-models (SDM) library, which provides a common in-context learning interface and also includes TabICLv2, Google’s TabFM, and KumoRelational. The GPU-oriented library supports preprocessing, ensembling, and many-class prediction. It requires Python 3.11+ and PyTorch 2.7+, with examples designed for CUDA GPUs.

The models were pretrained entirely on synthetic tables generated from Structural Causal Models, including simulated missing values, high-cardinality categories, heavy-tailed targets, and duplicate conflicts. NVIDIA reports that Kumo Tabular ranked first on TabArena with an Elo score of 1950 and led results across several other benchmarks. Its weights use the OpenMDW-1.1 license, which permits commercial use; the SDM code is released under Apache-2.0.

Written by Sakhanda Wire AI from reporting by MarkTechPost. Facts, quotes and full context are in the original article below.

Originally published by MarkTechPost on

Read the original on MarkTechPost ↗

The reporting and the image belong to MarkTechPost and are credited and linked above; the text on this page is Sakhanda Wire's own summary.

← Back to the news

More stories

All the latest news