Kumo Relational

Description

Kumo Relational is an NVIDIA-owned structured-data foundation model for prediction over relational, multi-table data. It uses declared schemas, relationships, entity tables, fact tables, context rows, and prediction rows without requiring callers to flatten connected tables into a single feature table.

The Kumo Relational NVIDIA Inference Microservice (NIM) accepts JSON arrays-format relational payloads through a structured-data API, validates the task, schema, context, and prediction contract, and invokes the Kumo Relational v1.0 driver for synchronous inference.

This model is licensed for non-commercial research or evaluation purposes only.

License and Terms of Use:

GOVERNING TERMS: The NIM container is governed by the NVIDIA Software License Agreement and the Product-Specific Terms for NVIDIA AI Products; and the use of the model is governed by the NVIDIA License.

Deployment Geography:

Global

Use Case:

Use Case: Kumo Relational is designed for binary classification, multiclass classification, regression, forecasting, and temporal link prediction over structured relational data. It supports one-shot prediction and session-based workflows that reuse the same model, task, schema, and context across prediction requests.

Release Date:

NGC 08/18/2026

Model Architecture:

Architecture Type: Transformer Network Architecture: Kumo Relational v1.0 uses a table-invariant row encoder and a Relational Graph Transformer to exchange information across connected tables and context labels. Total Parameters: Undisclosed

Input:

Input Types: Tabular Input Formats: JSON arrays-format relational tables Input Parameters: One Dimensional (1D) Other Input Properties: Requests declare an instance table, optional related tables, relationships, task metadata, labeled context rows, and prediction rows. Binary classification, multiclass classification, regression, and forecasting require exactly one entity table. Temporal link prediction requires source and target entity tables.

Output:

Output Types: Tabular prediction Output Format: JSON prediction results Output Parameters: One Dimensional (1D) Other Output Properties: Task-dependent outputs include predictions, probabilities, scores, rankings, quantiles, embeddings, and single-row diagnostic explanations.

Our AI models are designed and/or optimized to run on NVIDIA GPU-accelerated systems. By leveraging NVIDIA's hardware (e.g. GPU cores) and software frameworks (e.g., CUDA libraries), the model achieves faster training and inference times compared to CPU-only solutions.

Software Integration:

Runtime Engines: Kumo Relational v1.0 driver

Supported Hardware:

• NVIDIA A10G Tensor Core GPU
• NVIDIA H100 Tensor Core GPU
• NVIDIA L4 Tensor Core GPU
• NVIDIA L40S GPU

Operating System: Linux

The integration of foundation and fine-tuned models into AI systems requires additional testing using use-case-specific data to ensure safe and effective deployment. Following the V-model methodology, iterative testing and validation at both unit and system levels are essential to mitigate risks, meet technical and functional requirements, and ensure compliance with safety and ethical standards before deployment.

Model Version(s)

v1.0

Training, Testing, and Evaluation Datasets:

Training Dataset

Data Modality: Tabular Text Training Data Size: Undisclosed Data Collection Method by dataset: [Hybrid: Automated, Synthetic] Labeling Method by dataset: [Automated] Properties: Kumo Relational Foundation Model was pretrained on a mixture of publicly available real-world relational databases and synthetic data. The published whitepaper states that no private enterprise data was used for training. Historical labels are sampled and attached to relational subgraphs for in-context supervision. Exact dataset identities, size bucket, review cadence, and complete provenance remain subject to data-owner confirmation.

Testing Dataset

Data Collection Method by dataset: [Hybrid: Automated, Human] Labeling Method by dataset: [Hybrid: Automated, Human] Properties: Release testing uses the structured-data NIM benchmark harness with deterministic and pinned external benchmark inputs, including RelBench and SALT task families. Final release-candidate measurements remain subject to the official SWQA report.

Evaluation Dataset

Evaluation Benchmark Score: Undisclosed Data Collection Method by dataset: [Hybrid: Automated, Human] Labeling Method by dataset: [Hybrid: Automated, Human] Properties: The public Kumo Relational evaluation covers 30 predictive tasks from seven public RelBench datasets and states that Kumo Relational was not trained or tuned on those datasets or tasks. Final NIM release-candidate scores remain subject to the official SWQA report.

Inference

Acceleration Engine: Kumo Relational v1.0 driver Supported Tasks: Binary classification, multiclass classification, regression, forecasting, and temporal link prediction API Endpoints: POST /v1/predictions, POST /v1/sessions, POST /v1/sessions/{session_id}/predictions, and DELETE /v1/sessions/{session_id} Test Hardware:

• NVIDIA A10G Tensor Core GPU
• NVIDIA H100 Tensor Core GPU
• NVIDIA L4 Tensor Core GPU
• NVIDIA L40S GPU

Ethical Considerations

NVIDIA believes Trustworthy AI is a shared responsibility and we have established policies and practices to enable development for a wide array of AI applications. Developers should work with their internal developer team to ensure these software components meet requirements for the relevant industry and use case and address unforeseen product misuse.

Users are responsible for model inputs and outputs. Users are responsible for ensuring safe integration of this model, including implementing guardrails as well as other safety mechanisms, prior to deployment.

Enterprise Support

Support for NVIDIA Kumo Relational is provided through NVIDIA's enterprise support channels for customers with an entitlement that covers this product. Support scope, response targets, and lifecycle are defined by your NVIDIA AI Enterprise agreement, not by this model card.

• Support link: NVIDIA Enterprise Support Portal
• Entitlement and coverage: NVIDIA Enterprise Support
• Support policy and lifecycle: NVIDIA AI Enterprise support documentation

Get Help

• Enterprise customers: open a case in the NVIDIA Enterprise Support Portal.
• All users: ask questions in the NVIDIA Developer Forums.
• Product documentation: NVIDIA AI Enterprise documentation.
• Security issues: report model quality, risk, or security vulnerabilities here.