> 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.

# What are embedding outputs?

> Generate and retain learned entity representations

Embeddings are learned vectors representing graph entities. You can use them for clustering, segmentation, retrieval, or as features for another model.

Select `PredictionArtifactType.EMBEDDINGS` in the prediction output configuration. Filesystem and object-storage connectors also support saving predictions and embeddings together. ID-GNN does not support embedding output.

See [Generate and Save Embeddings](/kumo-engine/1.0.0/user-code-guide/advanced-prediction-configurations#generate-and-save-embeddings) for task-specific options and storage examples.

To supply vectors computed by another model as input features, see [External Column Embeddings](/faq/kumo-engine/data-graphs/how-can-i-use-an-external-model-for-embedding-a-column).