> 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.

# Supported Models

> Model support in the current Kumo Relational Client and where to find local inference guides.

Kumo Relational Client supports the `kumo-relational` model through
a deployed Kumo Relational NIM. Use `client.relational(graph)` to create a model
handle and submit prediction requests.

For local tabular inference, use SDM. `RelationalClient` does not expose a
tabular prediction API.

| Workflow                                                                            | Documentation                                                          |
| ----------------------------------------------------------------------------------- | ---------------------------------------------------------------------- |
| Run relational predictions through a NIM endpoint                                   | [Kumo Relational NIM](/kumo-relational/about-kumo-relational/overview) |
| Load model weights and run inference directly with SDM, including a tabular example | [SDM User Guide](/structured-data-models/user-guide)                   |

See the [client repository](https://github.com/kumo-ai/kumo-relational-client)
and [Python API Reference](/kumo-relational/reference/python-api-reference) for the
current public interfaces.