> For clean Markdown of any page, append .md to the page URL.
> For a complete documentation index, see https://docs.nvidia.com/run-ai/self-hosted/v3.0/reference/rest-api/endpoints/workloads/inferences/llms.txt.
> For AI client integration (Claude Code, Cursor, etc.), connect to the MCP server at https://docs.nvidia.com/run-ai/_mcp/server.

# Inferences

Inference workloads deploy trained models into a production environment to generate predictions from live data. These workloads are prioritized over Trainings and Workspaces during scheduling. NVIDIA Run:ai Inference workloads support auto-scaling to maintain service-level agreements (SLAs) by dynamically adjusting resources as demand changes.


## API Docs

- Inferences [Create an inference.](https://docs.nvidia.com/run-ai/self-hosted/v3.0/reference/rest-api/endpoints/workloads/inferences/create-inference-1.md)
- Inferences [Get inference data.](https://docs.nvidia.com/run-ai/self-hosted/v3.0/reference/rest-api/endpoints/workloads/inferences/get-inference.md)
- Inferences [Delete an inference.](https://docs.nvidia.com/run-ai/self-hosted/v3.0/reference/rest-api/endpoints/workloads/inferences/delete-inference.md)
- Inferences [Update inference spec.](https://docs.nvidia.com/run-ai/self-hosted/v3.0/reference/rest-api/endpoints/workloads/inferences/update-inference-spec.md)
- Inferences [Get inference metrics data.](https://docs.nvidia.com/run-ai/self-hosted/v3.0/reference/rest-api/endpoints/workloads/inferences/get-inference-workload-metrics.md)
- Inferences [Get inference pod's metrics data.](https://docs.nvidia.com/run-ai/self-hosted/v3.0/reference/rest-api/endpoints/workloads/inferences/get-inference-workload-pod-metrics.md)

## OpenAPI Specification

The raw OpenAPI 3.1 specification for this API is available at:
- [OpenAPI JSON](https://docs.nvidia.com/run-ai/self-hosted/v3.0/reference/rest-api/endpoints/workloads/inferences/openapi.json)
- [OpenAPI YAML](https://docs.nvidia.com/run-ai/self-hosted/v3.0/reference/rest-api/endpoints/workloads/inferences/openapi.yaml)