---
title: "NVIDIA Kumo Relational Outperforms Supervised Machine Learning on Relational Data, No Training Required"
description: "NVIDIA Kumo Relational is a pre-trained relational foundation model that makes predictions directly on enterprise databases without requiring task-specific training, feature engineering, or table flattening. Given a predictive query, it gathers up to 10,000 historical examples as context and produces outcomes such as churn or demand in a single forward pass. Across 41 tasks on 15 databases, it outperforms supervised relational deep learning models and gradient-boosted tree pipelines; on the SAP SALT benchmark it surpasses AutoGluon ensembles by 8% in mean reciprocal rank. The model scales to databases with over 500 billion rows and is available through NVIDIA’s open-source structured-data-models library or as an NVIDIA NIM. It reasons within tables using column and row attention and across tables using graph attention over foreign keys, avoiding quadratic cost."
---

# NVIDIA Kumo Relational Outperforms Supervised Machine Learning on Relational Data, No Training Required

[Hugging Face](https://yomu.fyi/company/hugging-face) · Joshua Przybylko, Jure Leskovec, Matthias Fey, Federico Lopez, Valter Hudovernik, Akihiro Nitta, Ramona Bendias, Ardrian Wong · Oct 21, 2026

**Type:** Announcement

## Summary

NVIDIA Kumo Relational is a pre-trained relational foundation model that makes predictions directly on enterprise databases without requiring task-specific training, feature engineering, or table flattening. Given a predictive query, it gathers up to 10,000 historical examples as context and produces outcomes such as churn or demand in a single forward pass. Across 41 tasks on 15 databases, it outperforms supervised relational deep learning models and gradient-boosted tree pipelines; on the SAP SALT benchmark it surpasses AutoGluon ensembles by 8% in mean reciprocal rank. The model scales to databases with over 500 billion rows and is available through NVIDIA’s open-source structured-data-models library or as an NVIDIA NIM. It reasons within tables using column and row attention and across tables using graph attention over foreign keys, avoiding quadratic cost.

## Context

Traditional predictive modeling on relational databases requires weeks of feature engineering, task-specific training, and flattening of tables into single rows, which loses structural signals and demands significant effort.

## Approach / What changed

Kumo Relational is a pre-trained relational foundation model that uses automatic context generation from historical database states, reasons within tables via column and row attention, and reasons across tables via graph attention over primary-key/foreign-key relationships, interleaved with cross-sample attention. It is pre-trained on synthetic relational data and adapts through in-context learning with up to 10,000 examples, requiring no task-specific training.

## Takeaways

- Kumo Relational achieves superior accuracy compared to supervised learning pipelines on relational data without any task-specific training, using only in-context examples from the database.
- The model operates directly on relational databases, preserving foreign-key relationships and timestamps, which avoids the information loss associated with flattening tables into single rows.
- It scales to databases with over 500 billion rows via memory-mapped I/O at up to 5 GB/s and can be deployed as a GPU library or as a containerized NVIDIA NIM service.

## FAQ

### Does Kumo Relational require training on my data?

No, it uses in-context learning with up to 10,000 historical examples from your database, so no task-specific training is needed. Fine-tuning is optional for maximum accuracy.

### How does Kumo Relational handle multiple related tables?

It reasons within each table using column and row attention, then across tables using graph attention over primary-key/foreign-key relationships, avoiding the need to flatten data.

### Can agents use Kumo Relational for predictions?

Yes, it offers a Predictive Query Language (PQL) interface that agents can call to ask predictive questions, such as which customer will churn, and act on the results.

**Tags:** [AI](https://yomu.fyi/topic/ai), [Foundation Models](https://yomu.fyi/topic/foundation-models), [Machine Learning](https://yomu.fyi/topic/machine-learning), [NVIDIA](https://yomu.fyi/topic/nvidia), [Relational Data](https://yomu.fyi/topic/relational-data)

- Source: [Hugging Face](https://huggingface.co/blog/nvidia/kumo-relational)
- Source URL: https://huggingface.co/blog/nvidia/kumo-relational
- Ingested by Yomu: 2026-10-21T21:00:13.552Z

[Read original post](https://huggingface.co/blog/nvidia/kumo-relational)