nemo_relay.adaptive
nemo_relay.adaptive
nemo_relay.adaptive
Generated from python/nemo_relay/adaptive.pyi.
Module nemo_relay.adaptive.
Adaptive plugin configuration helpers.
ConfigDiagnostic(TypedDict)One adaptive validation diagnostic.
ConfigReport(TypedDict)Validation report for adaptive configuration.
ConfigPolicyPolicy for unsupported adaptive configuration.
to_dictSerialize this policy to the canonical JSON object shape.
BackendSpecAdaptive state backend selection.
in_memoryReturn an in-memory adaptive backend spec.
redisReturn a Redis adaptive backend spec.
to_dictSerialize this backend spec to the canonical JSON object shape.
StateConfigAdaptive state configuration.
to_dictSerialize this state config to the canonical JSON object shape.
TelemetryConfigBuilt-in adaptive telemetry subscriber settings.
to_dictSerialize this telemetry config to the canonical JSON object shape.
AdaptiveHintsConfigBuilt-in adaptive hints injection settings.
to_dictSerialize this adaptive-hints config to the canonical JSON object shape.
ToolParallelismConfigBuilt-in adaptive tool scheduling settings.
to_dictSerialize this tool-parallelism config to the canonical JSON object shape.
AcgStabilityThresholdsPrompt-stability classification thresholds for ACG.
to_dictSerialize these ACG stability thresholds to the canonical JSON object shape.
AcgConfigAdaptive Cache Governor settings.
to_dictSerialize this ACG config to the canonical JSON object shape.
AdaptiveConfigCanonical config document for the top-level adaptive component.
to_dictSerialize this adaptive config to the canonical JSON object shape.
ComponentSpecTop-level adaptive component wrapper.
to_dictSerialize this component to the canonical plugin shape.
AdaptiveRuntimeHosted adaptive runtime wrapper used by external framework integrations.
AdaptiveRuntimeCreate an instance.
registerRegister the configured adaptive features with NeMo Relay.
deregisterDeregister previously registered adaptive features.
shutdownDeregister the runtime and release its owned resources.
wait_for_idleBlock until the adaptive telemetry drain has processed pending work.
reportReturn the validation report associated with this runtime.
bind_scopeBind this runtime’s ACG request rewriting to an active scope.
build_cache_request_factsBuild cache-diagnostics facts for an annotated request.
validate_configValidate an adaptive config document without constructing a runtime.
build_cache_telemetry_eventBuild one canonical cache telemetry event from usage plus request facts.
set_latency_sensitivitySet a request-local latency-sensitivity hint.
ADAPTIVE_PLUGIN_KIND