Powering Predictions in Snowflake Intelligence with KumoRFM

We’re excited to share that Kumo is partnering with Snowflake to make Predictive Intelligence a native part of the Snowflake Intelligence experience.

Together, we’re making it possible for anyone, from sales and marketing teams to operations and finance, to ask predictive questions in plain language and get instant, actionable answers from their data.

This partnership marks an exciting step forward in our mission: to make advanced machine learning effortless and accessible to every organization.

A New Way to Make Predictions with Your Data

With Snowflake Intelligence, teams can now use natural language to explore data across sources, from internal CRM tables to third-party contextual streams, all within Snowflake’s secure environment.

Kumo extends that vision by adding the ability to predict what happens next.Imagine asking:

• “Which deals are likely to close in 30 days?”
• “Which products risk stockout next quarter?”
• “Which transactions show the highest probability of fraud?”

and getting predictions immediately. No ML expertise required.

The Prediction Engine: Kumo Relational Foundation Model (KumoRFM)

The prediction engine is KumoRFM, a Relational Foundation Model built for predictive analytics on structured data. Unlike traditional machine learning models, KumoRFM leverages in-context learning, similar to large language models, to adapt to new datasets and prediction tasks at inference time. This entirely new approach to machine learning removes the need for extensive feature engineering or bespoke model training, while also delivering superior performance. On benchmarks such as RelBench, KumoRFM outperforms standard supervised learning methods by 2–8%, with an additional 10–30% boost when fine-tuned [1].

At its core, KumoRFM uses a Graph Transformer-based architecture for relational deep learning, pre-trained on volumes of public and synthetic relation data, enabling training-free predictions. Kumo RFM fundamentally streamlines the process of generating AI predictions from relational data, making it simple to integrate those predictions directly into agentic workflows.

What Does It Look Like?

In this demo, we’ll show you a Sales Intelligence Agent, leveraging Snowflake Cortex Agent API, Snowflake Semantic Models, Langchain-Snowflake, and KumoRFM.

Getting Started

Fill out the form to contact our team to enable a trial of the KumoRFM Native Application in your Snowflake account so you can get started building your own predictive agents securely within Snowflake.

So, what do you want to predict today?

References: [1] Kumo.ai (2025) KumoRFM: A foundation model for in-context learning on relational data

Join our community on Discord.

Connect with developers and data professionals, share ideas, get support, and stay informed about product updates, events, and best practices.

Join the discussion