NVIDIA has introduced Kumo Relational, a relational foundation model designed to make predictions directly on enterprise databases without task-specific training, feature engineering, or table flattening. According to the company, the model outperformed supervised relational deep learning models and expert-built gradient boosting tree pipelines across 41 tasks spanning 15 databases. NVIDIA also said that, in the SAP SALT enterprise benchmark, Kumo Relational delivered an 8% higher Mean Reciprocal Rank (MRR) than the AutoGluon ensemble approach. The company added that the model can scale to databases with more than 500 billion rows. Kumo Relational is being offered through NVIDIA’s open-source GPU-native structured data model library and via NVIDIA NIM services. The release positions the model as a tool for enterprises seeking database prediction capabilities without the usual preprocessing and task-level model training steps.
NVIDIA has released Kumo Relational, a relational foundation model that can make predictions directly on enterprise databases without task-specific training, feature engineering, or table flattening.
According to the announcement, the model outperformed supervised relational deep learning models and expert-designed gradient boosting tree pipelines across 41 tasks on 15 databases.
Benchmark results
In the SAP SALT enterprise benchmark, Kumo Relational posted an 8% higher Mean Reciprocal Rank (MRR) than the AutoGluon ensemble method.
Scale and availability
NVIDIA said the model can scale to databases with more than 500 billion rows. It is available through NVIDIA’s open-source GPU-native structured data model library or through NVIDIA NIM.
This article was originally published by Bit.Fan. For more cryptocurrency news and market insights, visit www.bit.fan. Disclaimer:
The market information, project data, and third-party content displayed on this platform are for industry information sharing only and do not constitute any form of investment advice or return commitment.
Cryptocurrency trading carries high risks. Users should fully assess their risk tolerance and make independent decisions. All profits, losses, and legal responsibilities are borne by the users themselves.
