TabTune: A Unified Library for Inference and Fine-Tuning Tabular Foundation Models

TL;DR

TabTune is a comprehensive library that standardizes the workflow for tabular foundation models, supporting multiple models, adaptation strategies, and evaluation metrics to facilitate deployment and benchmarking.

Contribution

It introduces a unified, extensible framework that automates preprocessing, manages heterogeneity, and evaluates models for tabular data learning.

Findings

Supports seven state-of-the-art models with various adaptation strategies

Automates preprocessing and manages architectural heterogeneity

Includes modules for performance, calibration, and fairness evaluation

Abstract

Tabular foundation models represent a growing paradigm in structured data learning, extending the benefits of large-scale pretraining to tabular domains. However, their adoption remains limited due to heterogeneous preprocessing pipelines, fragmented APIs, inconsistent fine-tuning procedures, and the absence of standardized evaluation for deployment-oriented metrics such as calibration and fairness. We present TabTune, a unified library that standardizes the complete workflow for tabular foundation models through a single interface. TabTune provides consistent access to seven state-of-the-art models supporting multiple adaptation strategies, including zero-shot inference, meta-learning, supervised fine-tuning (SFT), and parameter-efficient fine-tuning (PEFT). The framework automates model-aware preprocessing, manages architectural heterogeneity internally, and integrates evaluation modules for performance, calibration, and fairness. Designed for extensibility and reproducibility, TabTune enables consistent benchmarking of adaptation strategies of tabular foundation models.