Module runtime | NVIDIA NeMo Fabric

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

Runtime invocation helpers.

Structs

AdapterInvocation: One invocation against an initialized adapter runtime.
ArtifactManifest: Manifest of run artifacts.
ArtifactRef: Reference to one artifact.
EnvironmentHandle: Resolved execution environment context.
ErrorInfo: Normalized error metadata.
FabricEvent: NeMo Fabric lifecycle or progress event.
InvocationHandle: One request sent to a runtime.
OpenAiChatCompletionChunk: OpenAI Chat Completions chunk accepted by the native streaming profile.
OpenAiChatCompletionChunkChoice: One choice within an OpenAI Chat Completions streaming chunk.
OpenAiChatCompletionChunkDelta: Incremental assistant message fields carried by one OpenAI choice.
OpenAiStreamInvocation: One adapter-native OpenAI streaming invocation.
OpenAiStreamSink: Adapter-facing stream sink with invocation identity generated by NVIDIA NeMo Fabric.
OpenAiStreamTransport: SDK-owned loopback transport for one native OpenAI streaming invocation.
RunRequest: A request passed to a NeMo Fabric-managed harness runtime.
RunResult: Result from a NeMo Fabric-managed harness invocation.
RunUsage: Normalized invocation usage exposed to NeMo Fabric consumers.
RuntimeContext: Context generated for one invocation of a started runtime.
RuntimeHandle: Active or resumable harness runtime.
RuntimeTelemetryContext: Runtime telemetry config passed to adapters.
TelemetryRef: Reference to telemetry emitted by Relay or another configured telemetry path.

ErrorStage: NeMo Fabric lifecycle stage associated with an error.
OpenAiChatCompletionChunkObject: Exact OpenAI object discriminator accepted by the v1 chunk profile.
OpenAiStreamHost: Supported native-streaming listener host.
OpenAiStreamProfile: Supported OpenAI-compatible chunk profile.
OpenAiStreamProtocolVersion: Supported southbound native-streaming protocol version.
OpenAiStreamRecord: One correlated NDJSON record on the adapter-native stream channel.
RunStatus: Runtime completion status.

invoke_openai_stream: Invoke a started harness runtime and pass through native OpenAI chat-completion chunks.
invoke_runtime: Invoke a started harness runtime.
prepare_environment: Resolve or attach to the execution environment context for a run plan.
run_plan: Invoke a NeMo Fabric run plan.
start_runtime: Start or connect to a harness runtime.
stop_runtime: Stop or detach from a harness runtime.