Memory Efficient Tabular Foundation Models
arXiv:2607.27546
The paper studies how to reduce the memory footprint of tabular foundation models like TabPFN using compression techniques, achieving up to 7.6× memory savings with little performance loss for practical deployment.
Abstract
Tabular Foundation Models, such as TabPFN, have received a large amount of recent attention due to their performance on in-context tabular machine learning tasks, which often exceeds classical baselines. However, practical deployment considerations of these models has received less attention. In this paper we investigate the memory requirements for these models. We demonstrate that employing model compression approaches can enable memory reductions of up to 7.6 with similar levels of performance, reducing deployment requirements by nearly 87%. Our work provides insight to practitioners seeking efficient deployment of these models in practical settings.
12 pages, 3 figures Accepted at FMSD @ ICML 2026