Unlock Predictive AI in Snowflake

Build predictive AI
models in Snowflake

Kumo helps data scientists build highly accurate predictive AI models that predict user and customer behavior by combining graph learning over enterprise data and generative AI models trained on public data – all entirely within Snowflake.

• Graph transformers eliminate feature engineering and create predictive AI models from your Snowflake data.

• A graph neural network brings deep learning across the entirety of relational data inside Snowflake.

• Open-source and commercial LLMs like Cortex unlock unstructured data for even better predictions.

• Kumo’s AI learns directly on multiple Snowflake tables, and writes back predictions as new Snowflake tables.

• Relational data stored in Snowflake tables hold value that is extremely hard to unlock using traditional methods. To address this challenge, Kumo represents Snowflake tables as a graph and uses graph machine learning directly on raw data tables.

• Kumo’s graph neural network uses the natural relational structure of Snowflake tables with additional context about unstructured data from LLMs to maximize signal and improve the accuracy and performance of predictive models.

• Kumo provides detailed model evaluation, model explainability, MLOps, and more to build trust and monitor models in production including explanations down to the raw data. Better understand how various columns affect every prediction, how inputs contribute to a specific prediction, and relevant trends hidden in raw input data.

Benefits of Kumo
as a Snowflake Native App

Currently in private preview through Snowflake Marketplace, joint Snowflake and Kumo customers can access Kumo as a Snowflake Native App using a single login to reliably and securely create predictive models without sensitive data ever leaving their Snowflake environment.