> For clean Markdown of any page, append .md to the page URL. > For a complete documentation index, see https://docs.nvidia.com/sdgm/llms.txt. > For AI client integration (Claude Code, Cursor, etc.), connect to the MCP server at https://docs.nvidia.com/sdgm/_mcp/server. # Kumo Engine Overview > An open source foundation for learning from relational data, building predictive models, and putting predictions to work. [](#engine-animation) Animation source **Kumo Engine is an accelerator for context generation over structured data, designed for teams building and running predictive models in production at scale.** Like DuckDB and Polars for analytics, it is purpose-built for preparing the data models consume, combining fast context retrieval with time-correct sampling for training and batch inference. At two to four hops, Kumo Engine’s warm-cache runs generated context for one million entities [10 to 48× faster](/kumo-engine/1.0.0/resources/benchmarks#scale-and-model-input-performance) than prepared DuckDB and 9.5 to 22× faster than Polars on GPU. Its Rust-based engine and NVMe storage delivered this performance on a 2.6 TiB graph index using a machine with [just 1 TiB of RAM](/kumo-engine/1.0.0/resources/benchmarks#scale-and-model-input-performance). Kumo Engine gives developers and coding agents a powerful foundation for ingestion and large scale prediction. ## What You Can Build #### [Fraud](/kumo-engine/1.0.0/user-code-guide/code-examples/linux-deployment/train-a-classification-model) Flag suspicious transactions with a classifier trained on money-laundering examples. #### [Multiclass Classification](/kumo-engine/1.0.0/get-started-with-kumo-engine/workflows-and-solutions/predictive-query-examples#predict-the-next-purchase-channel) Predict a category, such as the channel a customer will use for their next purchase. #### [Forecasting](/kumo-engine/1.0.0/user-code-guide/code-examples/linux-deployment/forecasting) Forecast customer order counts across future time windows. #### [Regression](/kumo-engine/1.0.0/get-started-with-kumo-engine/workflows-and-solutions/predictive-query-examples#forecast-an-active-customers-spending) Estimate a numerical value, such as how much a customer will spend next month. #### [Link Prediction](/kumo-engine/1.0.0/user-code-guide/code-examples/linux-deployment/product-recommendations) Rank the products each customer is likely to buy for personalized recommendations. #### [Embeddings](/kumo-engine/1.0.0/user-code-guide/advanced-prediction-configurations#generate-and-save-embeddings) Generate customer and product vectors for downstream analysis. Different tasks can reuse the same connected data while defining their own targets, training examples, and evaluation criteria. Check out more use cases and examples in [Predictive Query Examples](/kumo-engine/1.0.0/get-started-with-kumo-engine/workflows-and-solutions/predictive-query-examples). ## Why Build on Kumo ### Use Relationships and History as Model Inputs A customer's orders, the products in those orders, and the timing of each purchase provide context beyond a single customer row. A total-spend feature, for example, leaves out which products a customer bought and in what order. Kumo's graph-learning workflow learns from the connected records you select, using the history available at each prediction time. ### Reuse the Data Foundation as Experiments Change Test whether more context, a different encoder, or a larger model helps your task by changing a model plan. Kumo handles context retrieval and sampling for each run, while you compare configurations using the same training examples and validation splits. For a new prediction target, generate historical examples from the same connected data. ### Iterate with Coding Agents Ask your coding agent to compare two sampling depths within a fixed training budget. It can update the [model plan](/kumo-engine/1.0.0/user-code-guide/train-a-model/model-plan-guide), run both configurations, and use [training diagnostics](/kumo-engine/1.0.0/user-code-guide/train-a-model/training-diagnostics) to summarize validation scores, learning curves, and runtime. Kumo handles the data loading and temporal sampling, so the agent can move directly from a modeling question to an experiment. Use the results to choose the next change, such as testing a different encoder or a smaller model. ## How It Works **DiskGraph**, Kumo's data engine built in Rust, indexes records and relationships on local SSD. It efficiently retrieves connected records and their features as model-ready batches, applying temporal constraints where relevant. The learning runtime consumes those batches for training and prediction; the [Model Plan Guide](/kumo-engine/1.0.0/user-code-guide/train-a-model/model-plan-guide) covers configuration. A **predictive query** specifies what to predict and for which entities, including a time window for temporal tasks. For example, **“Which customers will buy in the next 90 days?”** defines a purchase-propensity task with customers as the entities and purchases in that future window as the outcome. See [Predictive Queries](/kumo-engine/1.0.0/user-code-guide/train-a-model/predictive-queries) for the query and its Python workflow. ## Your Environment, Your Models, Your Control Run Kumo Engine on a supported Linux machine or in a Snowflake or Databricks notebook. Your credentials govern data access, you provision compute and storage, and model plans configure the modeling workflow. See the [setup guidance](/kumo-engine/1.0.0/get-started-with-kumo-engine/available-deployment-options) and [prerequisites](/kumo-engine/1.0.0/get-started-with-kumo-engine/prerequisites) for environment requirements. ## Start with One Useful Prediction #### [Choose Your Environment](/kumo-engine/1.0.0/get-started-with-kumo-engine/available-deployment-options) Compare Linux, Snowflake, and Databricks setup paths, then prepare your compute, storage, and data access. #### [Follow the User Code Guide](/kumo-engine/1.0.0/user-code-guide/set-up-a-graph/tables) Define source tables, then follow the guides for graphs, training, and prediction.