> For clean Markdown of any page, append .md to the page URL.
> For a complete documentation index, see https://docs.nvidia.com/nemo/fabric/dev/reference/api/rust-library-reference/nemo-fabric-core/runtime/llms.txt.
> For AI client integration (Claude Code, Cursor, etc.), connect to the MCP server at https://docs.nvidia.com/nemo/fabric/_mcp/server.

# Module runtime

> Runtime invocation helpers.

Generated from `cargo doc --no-deps -p nemo-fabric-core`.

Runtime invocation helpers.

## Structs

* [AdapterInvocation](/nemo/fabric/dev/reference/api/rust-library-reference/nemo-fabric-core/runtime/struct-adapterinvocation): One invocation against an initialized adapter runtime.
* [ArtifactManifest](/nemo/fabric/dev/reference/api/rust-library-reference/nemo-fabric-core/runtime/struct-artifactmanifest): Manifest of run artifacts.
* [ArtifactRef](/nemo/fabric/dev/reference/api/rust-library-reference/nemo-fabric-core/runtime/struct-artifactref): Reference to one artifact.
* [EnvironmentHandle](/nemo/fabric/dev/reference/api/rust-library-reference/nemo-fabric-core/runtime/struct-environmenthandle): Resolved execution environment context.
* [ErrorInfo](/nemo/fabric/dev/reference/api/rust-library-reference/nemo-fabric-core/runtime/struct-errorinfo): Normalized error metadata.
* [FabricEvent](/nemo/fabric/dev/reference/api/rust-library-reference/nemo-fabric-core/runtime/struct-fabricevent): NeMo Fabric lifecycle or progress event.
* [InvocationHandle](/nemo/fabric/dev/reference/api/rust-library-reference/nemo-fabric-core/runtime/struct-invocationhandle): One request sent to a runtime.
* [OpenAiChatCompletionChunk](/nemo/fabric/dev/reference/api/rust-library-reference/nemo-fabric-core/runtime/struct-openaichatcompletionchunk): OpenAI Chat Completions chunk accepted by the native streaming profile.
* [OpenAiChatCompletionChunkChoice](/nemo/fabric/dev/reference/api/rust-library-reference/nemo-fabric-core/runtime/struct-openaichatcompletionchunkchoice): One choice within an OpenAI Chat Completions streaming chunk.
* [OpenAiChatCompletionChunkDelta](/nemo/fabric/dev/reference/api/rust-library-reference/nemo-fabric-core/runtime/struct-openaichatcompletionchunkdelta): Incremental assistant message fields carried by one OpenAI choice.
* [OpenAiStreamInvocation](/nemo/fabric/dev/reference/api/rust-library-reference/nemo-fabric-core/runtime/struct-openaistreaminvocation): One adapter-native OpenAI streaming invocation.
* [OpenAiStreamSink](/nemo/fabric/dev/reference/api/rust-library-reference/nemo-fabric-core/runtime/struct-openaistreamsink): Adapter-facing stream sink with invocation identity generated by NVIDIA NeMo Fabric.
* [OpenAiStreamTransport](/nemo/fabric/dev/reference/api/rust-library-reference/nemo-fabric-core/runtime/struct-openaistreamtransport): SDK-owned loopback transport for one native OpenAI streaming invocation.
* [RunRequest](/nemo/fabric/dev/reference/api/rust-library-reference/nemo-fabric-core/runtime/struct-runrequest): A request passed to a NeMo Fabric-managed harness runtime.
* [RunResult](/nemo/fabric/dev/reference/api/rust-library-reference/nemo-fabric-core/runtime/struct-runresult): Result from a NeMo Fabric-managed harness invocation.
* [RunUsage](/nemo/fabric/dev/reference/api/rust-library-reference/nemo-fabric-core/runtime/struct-runusage): Normalized invocation usage exposed to NeMo Fabric consumers.
* [RuntimeContext](/nemo/fabric/dev/reference/api/rust-library-reference/nemo-fabric-core/runtime/struct-runtimecontext): Context generated for one invocation of a started runtime.
* [RuntimeHandle](/nemo/fabric/dev/reference/api/rust-library-reference/nemo-fabric-core/runtime/struct-runtimehandle): Active or resumable harness runtime.
* [RuntimeTelemetryContext](/nemo/fabric/dev/reference/api/rust-library-reference/nemo-fabric-core/runtime/struct-runtimetelemetrycontext): Runtime telemetry config passed to adapters.
* [ServiceHandle](/nemo/fabric/dev/reference/api/rust-library-reference/nemo-fabric-core/runtime/struct-servicehandle): Public identity and binding for one prepared or attached service.
* [ServiceReference](/nemo/fabric/dev/reference/api/rust-library-reference/nemo-fabric-core/runtime/struct-servicereference): Caller-supplied reference to an already-running service.
* [TelemetryRef](/nemo/fabric/dev/reference/api/rust-library-reference/nemo-fabric-core/runtime/struct-telemetryref): Reference to telemetry emitted by Relay or another configured telemetry path.

## Enums

* [ErrorStage](/nemo/fabric/dev/reference/api/rust-library-reference/nemo-fabric-core/runtime/enum-errorstage): NeMo Fabric lifecycle stage associated with an error.
* [OpenAiChatCompletionChunkObject](/nemo/fabric/dev/reference/api/rust-library-reference/nemo-fabric-core/runtime/enum-openaichatcompletionchunkobject): Exact OpenAI object discriminator accepted by the v1 chunk profile.
* [OpenAiStreamHost](/nemo/fabric/dev/reference/api/rust-library-reference/nemo-fabric-core/runtime/enum-openaistreamhost): Supported native-streaming listener host.
* [OpenAiStreamProfile](/nemo/fabric/dev/reference/api/rust-library-reference/nemo-fabric-core/runtime/enum-openaistreamprofile): Supported OpenAI-compatible chunk profile.
* [OpenAiStreamProtocolVersion](/nemo/fabric/dev/reference/api/rust-library-reference/nemo-fabric-core/runtime/enum-openaistreamprotocolversion): Supported southbound native-streaming protocol version.
* [OpenAiStreamRecord](/nemo/fabric/dev/reference/api/rust-library-reference/nemo-fabric-core/runtime/enum-openaistreamrecord): One correlated NDJSON record on the adapter-native stream channel.
* [RunStatus](/nemo/fabric/dev/reference/api/rust-library-reference/nemo-fabric-core/runtime/enum-runstatus): Runtime completion status.
* [ServiceOwnership](/nemo/fabric/dev/reference/api/rust-library-reference/nemo-fabric-core/runtime/enum-serviceownership): Ownership of a service exposed through NeMo Fabric.

## Functions

* [attach\_service](/nemo/fabric/dev/reference/api/rust-library-reference/nemo-fabric-core/runtime/fn-attach-service): Attach to a caller-owned long-lived service for a resolved run plan.
* [invoke\_openai\_stream](/nemo/fabric/dev/reference/api/rust-library-reference/nemo-fabric-core/runtime/fn-invoke-openai-stream): Invoke a started harness runtime and pass through native OpenAI chat-completion chunks.
* [invoke\_runtime](/nemo/fabric/dev/reference/api/rust-library-reference/nemo-fabric-core/runtime/fn-invoke-runtime): Invoke a started harness runtime.
* [prepare\_environment](/nemo/fabric/dev/reference/api/rust-library-reference/nemo-fabric-core/runtime/fn-prepare-environment): Resolve or attach to the execution environment context for a run plan.
* [prepare\_service](/nemo/fabric/dev/reference/api/rust-library-reference/nemo-fabric-core/runtime/fn-prepare-service): Prepare a Fabric-owned long-lived service for a resolved run plan.
* [release\_service](/nemo/fabric/dev/reference/api/rust-library-reference/nemo-fabric-core/runtime/fn-release-service): Release a Fabric-owned service or detach from a caller-owned service.
* [run\_plan](/nemo/fabric/dev/reference/api/rust-library-reference/nemo-fabric-core/runtime/fn-run-plan): Invoke a NeMo Fabric run plan.
* [start\_runtime](/nemo/fabric/dev/reference/api/rust-library-reference/nemo-fabric-core/runtime/fn-start-runtime): Start or connect to a harness runtime.
* [start\_runtime\_with\_service](/nemo/fabric/dev/reference/api/rust-library-reference/nemo-fabric-core/runtime/fn-start-runtime-with-service): Start a runtime connected to an active service.
* [stop\_runtime](/nemo/fabric/dev/reference/api/rust-library-reference/nemo-fabric-core/runtime/fn-stop-runtime): Stop or detach from a harness runtime.

> Runtime invocation helpers.

## Docs

- [Struct Adapter Invocation](https://docs.nvidia.com/nemo/fabric/dev/reference/api/rust-library-reference/nemo-fabric-core/runtime/struct-adapterinvocation.md): One invocation against an initialized adapter runtime.
- [Struct Artifact Manifest](https://docs.nvidia.com/nemo/fabric/dev/reference/api/rust-library-reference/nemo-fabric-core/runtime/struct-artifactmanifest.md): Manifest of run artifacts.
- [Struct Artifact Ref](https://docs.nvidia.com/nemo/fabric/dev/reference/api/rust-library-reference/nemo-fabric-core/runtime/struct-artifactref.md): Reference to one artifact.
- [Struct Environment Handle](https://docs.nvidia.com/nemo/fabric/dev/reference/api/rust-library-reference/nemo-fabric-core/runtime/struct-environmenthandle.md): Resolved execution environment context.
- [Struct Error Info](https://docs.nvidia.com/nemo/fabric/dev/reference/api/rust-library-reference/nemo-fabric-core/runtime/struct-errorinfo.md): Normalized error metadata.
- [Struct Fabric Event](https://docs.nvidia.com/nemo/fabric/dev/reference/api/rust-library-reference/nemo-fabric-core/runtime/struct-fabricevent.md): NeMo Fabric lifecycle or progress event.
- [Struct Invocation Handle](https://docs.nvidia.com/nemo/fabric/dev/reference/api/rust-library-reference/nemo-fabric-core/runtime/struct-invocationhandle.md): One request sent to a runtime.
- [Struct OpenAI Chat Completion Chunk](https://docs.nvidia.com/nemo/fabric/dev/reference/api/rust-library-reference/nemo-fabric-core/runtime/struct-openaichatcompletionchunk.md): OpenAI Chat Completions chunk accepted by the native streaming profile.
- [Struct OpenAI Chat Completion Chunk Choice](https://docs.nvidia.com/nemo/fabric/dev/reference/api/rust-library-reference/nemo-fabric-core/runtime/struct-openaichatcompletionchunkchoice.md): One choice within an OpenAI Chat Completions streaming chunk.
- [Struct OpenAI Chat Completion Chunk Delta](https://docs.nvidia.com/nemo/fabric/dev/reference/api/rust-library-reference/nemo-fabric-core/runtime/struct-openaichatcompletionchunkdelta.md): Incremental assistant message fields carried by one OpenAI choice.
- [Struct OpenAI Stream Invocation](https://docs.nvidia.com/nemo/fabric/dev/reference/api/rust-library-reference/nemo-fabric-core/runtime/struct-openaistreaminvocation.md): One adapter-native OpenAI streaming invocation.
- [Struct OpenAI Stream Sink](https://docs.nvidia.com/nemo/fabric/dev/reference/api/rust-library-reference/nemo-fabric-core/runtime/struct-openaistreamsink.md): Adapter-facing stream sink with invocation identity generated by NVIDIA NeMo Fabric.
- [Struct OpenAI Stream Transport](https://docs.nvidia.com/nemo/fabric/dev/reference/api/rust-library-reference/nemo-fabric-core/runtime/struct-openaistreamtransport.md): SDK-owned loopback transport for one native OpenAI streaming invocation.
- [Struct RunRequest](https://docs.nvidia.com/nemo/fabric/dev/reference/api/rust-library-reference/nemo-fabric-core/runtime/struct-runrequest.md): A request passed to a NeMo Fabric-managed harness runtime.
- [Struct RunResult](https://docs.nvidia.com/nemo/fabric/dev/reference/api/rust-library-reference/nemo-fabric-core/runtime/struct-runresult.md): Result from a NeMo Fabric-managed harness invocation.
- [Struct RunUsage](https://docs.nvidia.com/nemo/fabric/dev/reference/api/rust-library-reference/nemo-fabric-core/runtime/struct-runusage.md): Normalized invocation usage exposed to NeMo Fabric consumers.
- [Struct Runtime Context](https://docs.nvidia.com/nemo/fabric/dev/reference/api/rust-library-reference/nemo-fabric-core/runtime/struct-runtimecontext.md): Context generated for one invocation of a started runtime.
- [Struct Runtime Handle](https://docs.nvidia.com/nemo/fabric/dev/reference/api/rust-library-reference/nemo-fabric-core/runtime/struct-runtimehandle.md): Active or resumable harness runtime.
- [Struct Runtime Telemetry Context](https://docs.nvidia.com/nemo/fabric/dev/reference/api/rust-library-reference/nemo-fabric-core/runtime/struct-runtimetelemetrycontext.md): Runtime telemetry config passed to adapters.
- [Struct Service Handle](https://docs.nvidia.com/nemo/fabric/dev/reference/api/rust-library-reference/nemo-fabric-core/runtime/struct-servicehandle.md): Public identity and binding for one prepared or attached service.
- [Struct Service Reference](https://docs.nvidia.com/nemo/fabric/dev/reference/api/rust-library-reference/nemo-fabric-core/runtime/struct-servicereference.md): Caller-supplied reference to an already-running service.
- [Struct Telemetry Ref](https://docs.nvidia.com/nemo/fabric/dev/reference/api/rust-library-reference/nemo-fabric-core/runtime/struct-telemetryref.md): Reference to telemetry emitted by Relay or another configured telemetry path.
- [Enum Error Stage](https://docs.nvidia.com/nemo/fabric/dev/reference/api/rust-library-reference/nemo-fabric-core/runtime/enum-errorstage.md): NeMo Fabric lifecycle stage associated with an error.
- [Enum OpenAI Chat Completion Chunk Object](https://docs.nvidia.com/nemo/fabric/dev/reference/api/rust-library-reference/nemo-fabric-core/runtime/enum-openaichatcompletionchunkobject.md): Exact OpenAI object discriminator accepted by the v1 chunk profile.
- [Enum OpenAI Stream Host](https://docs.nvidia.com/nemo/fabric/dev/reference/api/rust-library-reference/nemo-fabric-core/runtime/enum-openaistreamhost.md): Supported native-streaming listener host.
- [Enum OpenAI Stream Profile](https://docs.nvidia.com/nemo/fabric/dev/reference/api/rust-library-reference/nemo-fabric-core/runtime/enum-openaistreamprofile.md): Supported OpenAI-compatible chunk profile.
- [Enum OpenAI Stream Protocol Version](https://docs.nvidia.com/nemo/fabric/dev/reference/api/rust-library-reference/nemo-fabric-core/runtime/enum-openaistreamprotocolversion.md): Supported southbound native-streaming protocol version.
- [Enum OpenAI Stream Record](https://docs.nvidia.com/nemo/fabric/dev/reference/api/rust-library-reference/nemo-fabric-core/runtime/enum-openaistreamrecord.md): One correlated NDJSON record on the adapter-native stream channel.
- [Enum RunStatus](https://docs.nvidia.com/nemo/fabric/dev/reference/api/rust-library-reference/nemo-fabric-core/runtime/enum-runstatus.md): Runtime completion status.
- [Enum Service Ownership](https://docs.nvidia.com/nemo/fabric/dev/reference/api/rust-library-reference/nemo-fabric-core/runtime/enum-serviceownership.md): Ownership of a service exposed through NeMo Fabric.
- [Function attach_service](https://docs.nvidia.com/nemo/fabric/dev/reference/api/rust-library-reference/nemo-fabric-core/runtime/fn-attach-service.md): Attach to a caller-owned long-lived service for a resolved run plan.
- [Function invoke_openai_stream](https://docs.nvidia.com/nemo/fabric/dev/reference/api/rust-library-reference/nemo-fabric-core/runtime/fn-invoke-openai-stream.md): Invoke a started harness runtime and pass through native OpenAI chat-completion chunks.
- [Function invoke_runtime](https://docs.nvidia.com/nemo/fabric/dev/reference/api/rust-library-reference/nemo-fabric-core/runtime/fn-invoke-runtime.md): Invoke a started harness runtime.
- [Function prepare_environment](https://docs.nvidia.com/nemo/fabric/dev/reference/api/rust-library-reference/nemo-fabric-core/runtime/fn-prepare-environment.md): Resolve or attach to the execution environment context for a run plan.
- [Function prepare_service](https://docs.nvidia.com/nemo/fabric/dev/reference/api/rust-library-reference/nemo-fabric-core/runtime/fn-prepare-service.md): Prepare a Fabric-owned long-lived service for a resolved run plan.
- [Function release_service](https://docs.nvidia.com/nemo/fabric/dev/reference/api/rust-library-reference/nemo-fabric-core/runtime/fn-release-service.md): Release a Fabric-owned service or detach from a caller-owned service.
- [Function run_plan](https://docs.nvidia.com/nemo/fabric/dev/reference/api/rust-library-reference/nemo-fabric-core/runtime/fn-run-plan.md): Invoke a NeMo Fabric run plan.
- [Function start_runtime](https://docs.nvidia.com/nemo/fabric/dev/reference/api/rust-library-reference/nemo-fabric-core/runtime/fn-start-runtime.md): Start or connect to a harness runtime.
- [Function start_runtime_with_service](https://docs.nvidia.com/nemo/fabric/dev/reference/api/rust-library-reference/nemo-fabric-core/runtime/fn-start-runtime-with-service.md): Start a runtime connected to an active service.
- [Function stop_runtime](https://docs.nvidia.com/nemo/fabric/dev/reference/api/rust-library-reference/nemo-fabric-core/runtime/fn-stop-runtime.md): Stop or detach from a harness runtime.