Expand description
Backend-neutral model execution contracts and algorithms.
This crate orchestrates opaque backend-native values. It deliberately has no dependency on an architecture implementation or execution backend.
Re-exports§
pub use automatic_support::residency_telemetry;pub use automatic_support::selected_text_bounded_requirement;pub use automatic_support::BoundedResidencySizingError;pub use backend::BarrierBackend;pub use backend::BroadcastBackend;pub use backend::CollectiveBackend;pub use backend::CommunicationBackend;pub use backend::EvenGatherBackend;pub use backend::FailureAgreementBackend;pub use backend::ParameterBackend;pub use backend::PointToPointBackend;pub use backend::RoleExactBoundaryValue;pub use backend::SubmissionBackend;pub use backend::SumReductionBackend;pub use backend::TransferBackend;pub use backend::UnevenGatherBackend;pub use backend::VariableAllToAllBackend;pub use cache::finalize_prompt_cache_shard;pub use cache::hash_prompt_cache_shard_payload;pub use cache::inspect_prompt_cache;pub use cache::prompt_cache_rank_path;pub use cache::resolve_prompt_cache_root;pub use cache::safe_prompt_cache_shard_path;pub use cache::validate_prompt_cache_manifest;pub use cache::CacheBlockLifecycle;pub use cache::CacheBlockStorage;pub use cache::CacheHostDemotionOperation;pub use cache::CacheHostPromotion;pub use cache::CacheIoAdmission;pub use cache::CacheIoCompletionDisposition;pub use cache::CacheIoExecutionState;pub use cache::CacheIoExecutionStateError;pub use cache::CacheIoOperation;pub use cache::CacheIoOperationKey;pub use cache::CacheIoOperationKind;pub use cache::CacheIoPreparation;pub use cache::CacheIoStartDisposition;pub use cache::CacheIoSubmission;pub use cache::CacheIoSubmissionOutcome;pub use cache::CacheIoTicket;pub use cache::CacheIoWorker;pub use cache::CacheIoWorkerError;pub use cache::CacheLayerResidencyReport;pub use cache::CacheLayerResidencyStats;pub use cache::CacheLifecycleError;pub use cache::CachePoolError;pub use cache::CachePoolLimits;pub use cache::CachePoolMembership;pub use cache::CachePoolReport;pub use cache::CachePoolReservation;pub use cache::CachePoolResource;pub use cache::CachePoolUsage;pub use cache::CacheResidencyConfigurationError;pub use cache::CacheResidencyPolicy;pub use cache::CacheResidencyPool;pub use cache::CacheResidencyReport;pub use cache::CacheResidencyTelemetry;pub use cache::CacheStorageError;pub use cache::CacheStoragePhase;pub use cache::LiveCacheBlockPublication;pub use cache::LiveCacheDiskPolicy;pub use cache::LiveCachePublicationError;pub use cache::MutableCacheTail;pub use cache::PagedCacheOptions;pub use cache::PromptCachePersistenceError;pub use cache::PromptCachePublication;pub use cache::ReversiblePromptCachePublication;pub use cache::CACHE_RESIDENCY_LAYER_REPORT_LIMIT;pub use cache::MAX_PROMPT_CACHE_SHARD_HEADER_BYTES;pub use cache::PROMPT_CACHE_CURRENT_FILE;pub use cache::PROMPT_CACHE_GENERATIONS_DIRECTORY;pub use communication::validate_communication_manifest_consensus;pub use communication::prepare_communication_realization;pub use communication::project_all_communication_manifests;pub use communication::project_communication_manifest;pub use communication::validate_compatible_communication_manifests;pub use communication::BoundaryDimensionContract;pub use communication::BoundaryFramingProtocol;pub use communication::BoundaryRoleContract;pub use communication::CommunicationCapabilities;pub use communication::CommunicationCapabilityError;pub use communication::CommunicationCompletionCapabilities;pub use communication::CommunicationCompletionPolicy;pub use communication::CommunicationGroupDescriptor;pub use communication::CommunicationGroupRequirements;pub use communication::CommunicationManifest;pub use communication::CommunicationManifestConsensusError;pub use communication::CommunicationManifestError;pub use communication::CommunicationOperation;pub use communication::CommunicationOperationRequirement;pub use communication::CommunicationPeerCounts;pub use communication::CommunicationRealizationError;pub use communication::CommunicationRouteDescriptor;pub use communication::CommunicationRouteId;pub use communication::CommunicationTensorLimits;pub use communication::CommunicationTopologyCapabilities;pub use communication::PreparedCommunicationRealization;pub use communication::RoleExactBoundaryContract;pub use communication::TopologyCommunicationPlan;pub use component::ComponentDomain;pub use component::ComponentGraph;pub use component::ComponentGraphError;pub use component::ComponentKind;pub use component::ComponentResidencyClass;pub use component::ComponentSpec;pub use composite::select_composite_realization;pub use composite::select_processor_execution;pub use composite::CompositeSelectionError;pub use composite::MediaPrimitiveCapabilities;pub use composite::ModalityProcessorRequirements;pub use composite::ProcessorExecutionRequirements;pub use composite::ProcessorPrimitive;pub use composite::ProcessorSelectionError;pub use composite::ProcessorSelectionRequest;pub use composite::SelectedCompositeRealization;pub use composite::SelectedProcessorExecution;pub use decision::FullyForcedTailDecision;pub use decision::PredictionDirective;pub use decision::SequentialDecision;pub use decision::SequentialDecisionBoundary;pub use decision::SequentialDecisionDiagnostic;pub use decision::SequentialDecisionDriver;pub use decision::SequentialDecisionError;pub use decision::SequentialDecisionMode;pub use decision::SequentialDecisionPlan;pub use decision::SequentialDecisionPlanError;pub use decision::SequentialDecisionSource;pub use decision::SequentialDecisionTraversal;pub use decision::SequentialSamplingState;pub use dense::DenseCacheMetrics;pub use dense::DenseDiskStreamReport;pub use dense::DenseExecutionGroupReport;pub use dense::DensePassCounterSnapshot;pub use dense::DensePassReport;pub use dense::DenseStreamTelemetry;pub use dense::DenseStreamTelemetryError;pub use dense::DenseTierResidencyReport;pub use draft::execute_draft_group;pub use draft::DraftGroupExecutionError;pub use draft::DraftStateTransaction;pub use execution::ExecutionGraph;pub use execution::ExecutionGraphError;pub use execution::ExecutionGroupId;pub use execution::ExecutionGroupSchedule;pub use execution::ExecutionGroupSpec;pub use execution::ExecutionScheduleError;pub use execution::ExecutionUnitAddress;pub use execution::ExecutionUnitLayout;pub use execution::ExecutionUnitLayoutError;pub use execution::ReadyGroupState;pub use execution_plan::execution_plan_quantization;pub use execution_plan::ExecutionPlanLoadError;pub use execution_plan::ResidencyDiagnostics;pub use expert::combine_routed_expert_tensor_parallel;pub use expert::combine_tensor_parallel_expert_outputs;pub use expert::plan_addressable_bank_bindings;pub use expert::reduce_routed_expert_tensor_parallel;pub use expert::reduce_tensor_parallel_expert_output;pub use expert::selected_addressable_parameter_bytes;pub use expert::AddressableBankBindingPlan;pub use expert::AddressableBankDistribution;pub use expert::AddressableBankMember;pub use expert::AddressableBankMemberError;pub use expert::AddressableBankMemberPlacement;pub use expert::AddressableBankParameter;pub use expert::AddressableBindingTransform;pub use expert::AddressableExpertRouteProvider;pub use expert::AddressableExpertRouteRequest;pub use expert::AddressableGatedProductBank;pub use expert::AddressableGroupedBank;pub use expert::ExpertRouteCombination;pub use expert::ExpertRouteExchange;pub use expert::ExpertRouteTensorMovement;pub use expert::IndexedMovement;pub use expert::ObservedExpertProvider;pub use expert::ObservedExpertProviderError;pub use expert::ParameterBankAcquisition;pub use expert::ResidentExpertProvider;pub use expert::RoutedExpertProvider;pub use expert::RoutedExpertRequest;pub use expert::RoutedExpertTensorParallelOutput;pub use expert::RoutedObservationPoint;pub use expert::TensorParallelRoutedExpertProvider;pub use generation::CausalModel;pub use generation::ConstrainedSampler;pub use generation::DefaultSampler;pub use generation::GenerationSampler;pub use generation::MirostatV2Sampler;pub use generation::PenaltyConfig;pub use generation::Sampler;pub use generation::SamplingBackend;pub use generation::SamplingConfigurationError;pub use generation::SpeculativeSampler;pub use generation::TokenDomain;pub use input::PreparedInputCacheIdentity;pub use input::PreparedInputCacheIdentityError;pub use input::PreparedInputInspector;pub use input::PreparedInputPart;pub use input::PreparedInputPayload;pub use input::PreparedModelInput;pub use inspection::observe_and_intervene;pub use inspection::observe_model_logits;pub use inspection::ActivationObserver;pub use inspection::NoopObserver;pub use inspection::RoutingObservation;pub use inspection::TargetStateCapture;pub use inspection::TargetStateCaptureError;pub use inspection::TargetStateTap;pub use layered::ArchitectureGroupKind;pub use layered::ArchitectureGroupPlacement;pub use layered::ArchitectureGroupTransport;pub use layered::ArchitectureMergeDestination;pub use layered::ArchitectureParallelSubgroup;pub use layered::ArchitectureParameters;pub use layered::CompositeLayeredTraversalHook;pub use layered::LayeredArchitecture;pub use layered::LayeredForwardState;pub use layered::LayeredPartitionInput;pub use layered::LayeredPartitionOutput;pub use layered::LayeredPipelineSchedule;pub use layered::LayeredPipelineScheduleError;pub use layered::LayeredTraversalHook;pub use layered::LayeredTraversalPoint;pub use layered::LayeredUnitAction;pub use layered::LayerwiseAcquireError;pub use layered::LayerwisePolicy;pub use layered::LayerwisePolicyForward;pub use layered::LayerwiseRuntime;pub use layered::LayerwiseRuntimeError;pub use layered::ParallelLayeredArchitecture;pub use layered::ParallelRoutedLayeredArchitecture;pub use layered::PartitionedLayeredArchitecture;pub use layered::ResidentRuntime;pub use layered::ResidentUnitWindow;pub use layered::ResidentUnitWindowError;pub use layered::RoutedLayeredArchitecture;pub use layered::StaticParameterVisitor;pub use layered::StaticParameterVisitorMut;pub use load_request::DraftingLoadRequest;pub use load_request::NormalizedLoadRequest;pub use load_request::NormalizedLoadRequestError;pub use load_request::ParallelLoadRequest;pub use load_request::ValidatedModelLoadRequest;pub use mechanism_synthesis::synthesize_replicated_text_capabilities;pub use mechanism_synthesis::BackendMechanismFacts;pub use mechanism_synthesis::ReplicatedTextMechanismSupport;pub use mechanism_synthesis::StateLifecycleCapabilities;pub use parallel::aligned_partition_units;pub use parallel::expand_linear_format_parameter_groups;pub use parallel::module_parameter_group;pub use parallel::partitioned_module_parameter_group;pub use parallel::partitioned_projection_group;pub use parallel::projection_parameter_group;pub use parallel::segmented_projection_group;pub use parallel::LocalModelLayout;pub use parallel::LocalTensorLayout;pub use parallel::MemberSharding;pub use parallel::ParallelModelInfo;pub use parallel::ParallelPlanError;pub use parallel::ParameterGroupSpec;pub use parallel::ParameterMemberSpec;pub use parallel::ParameterRole;pub use parallel::ProjectionSharding;pub use parallel::ShardingPolicy;pub use parallel::TensorPlacement;pub use parameter::bind_materialized_unit;pub use parameter::bind_materialized_unit_excluding;pub use parameter::bindings_from_recipe_set;pub use parameter::build_exact_replicated_text_bindings;pub use parameter::build_module_binding_plan;pub use parameter::materialize_bindings;pub use parameter::materialize_selected_bindings;pub use parameter::preflight_bindings;pub use parameter::select_bindings;pub use parameter::BindingPlan;pub use parameter::BindingPlanError;pub use parameter::MaterializedUnit;pub use parameter::ModuleBindingPlan;pub use parameter::ModuleBindingPlanError;pub use parameter::ParameterBindingTarget;pub use parameter::ParameterOrchestrationError;pub use parameter::PlannedBinding;pub use parameter::RecipeBindingError;pub use parameter::SelectedBindingPlan;pub use partition::validate_boundary_tensor_count;pub use partition::ArchitectureBoundary;pub use partition::ArchitectureBoundaryError;pub use partition::ArchitectureBoundaryValue;pub use partition::ArchitectureParameterDescription;pub use partition::ArchitectureParameterError;pub use partition::ArchitecturePartition;pub use partition::ArchitecturePartitionError;pub use partition::BoundaryTensorDimension;pub use partition::BoundaryTensorDtype;pub use partition::BoundaryTensorSpec;pub use partition::BoundaryWireSchema;pub use partition::LayeredPartitionBeginError;pub use partition::LayeredPartitionDriver;pub use partition::LayeredPartitionError;pub use partition::NoAuxiliaryBoundary;pub use partition::NoAuxiliaryBoundarySchema;pub use partition::OwnedParameterGroupSpec;pub use partition::ParameterGroupOwner;pub use partition::PartitionGroup;pub use partition::PartitionOwnership;pub use partition::PartitionState;pub use partition::PipelineActivationDtype;pub use partition::PipelineWireContract;pub use partition::ResolvedBoundaryTensorSpec;pub use partition::ResolvedBoundaryWireSchema;pub use partitioned_execution::CommunicationTensorMetadata;pub use partitioned_execution::DistributedExecutionPhase;pub use partitioned_execution::LayerwiseTraversalPartitionExecutor;pub use partitioned_execution::LayerwiseTraversalRuntime;pub use partitioned_execution::NoBoundaryTransport;pub use partitioned_execution::NoCommitAgreement;pub use partitioned_execution::NoOutputPublisher;pub use partitioned_execution::OpaqueBoundaryTransport;pub use partitioned_execution::OpaqueCommitAgreement;pub use partitioned_execution::OpaqueFailureAgreement;pub use partitioned_execution::OpaqueOutputPublisher;pub use partitioned_execution::PartitionBoundaryRoute;pub use partitioned_execution::PartitionBoundaryTransport;pub use partitioned_execution::PartitionCommitAgreement;pub use partitioned_execution::PartitionCommunication;pub use partitioned_execution::PartitionCommunicationAuthority;pub use partitioned_execution::PartitionExecutionError;pub use partitioned_execution::PartitionOutputAuthority;pub use partitioned_execution::PartitionOutputPublication;pub use partitioned_execution::PartitionOutputPublisher;pub use partitioned_execution::PartitionedExecutionPlan;pub use partitioned_execution::PartitionedGroupExecutor;pub use partitioned_execution::PartitionedTextExecution;pub use partitioned_execution::PartitionedTextRuntime;pub use partitioned_execution::PartitionedTraversalError;pub use partitioned_execution::PartitionedTraversalResult;pub use partitioned_execution::RealizedCommunicationGroup;pub use partitioned_execution::RealizedCommunicationRoute;pub use placement::place_addressable_member_bindings;pub use placement::place_weight_bindings;pub use placement::placement_selection;pub use placement::BindingPlacementError;pub use placement::PlacementPlan;pub use placement::PlacementPlanError;pub use placement::PlacementRank;pub use placement::ResolvedTensorPlacement;pub use placement::TensorSlice;pub use prefetch::BackgroundPrefetchWorker;pub use prefetch::BackgroundPrefetchWorkerError;pub use realtime::RealtimeCompletionAttachmentError;pub use realtime::RealtimeFrameExecutionError;pub use realtime::RealtimeFrameTransition;pub use realtime::RealtimeGenerationBranch;pub use realtime::RealtimeGenerationState;pub use realtime::RealtimeGenerationTransactionError;pub use realtime_executor::execute_realtime_frame;pub use realtime_executor::PreparedRealtimeFrameExecutor;pub use realtime_executor::PrepublicationRealtimeFrame;pub use realtime_executor::RealtimeCompletionCreationError;pub use realtime_executor::RealtimeDecisionExecution;pub use realtime_executor::RealtimeFrameCompletionMechanism;pub use realtime_executor::RealtimeFrameCoordinatorError;pub use realtime_executor::RealtimeFrameHostObserver;pub use realtime_executor::RealtimePrepublicationError;pub use realtime_executor::SubmittedRealtimeFrame;pub use realtime_ingress::MaterializedRealtimeInput;pub use realtime_ingress::RealtimeHostTokenMaterializer;pub use realtime_ingress::RealtimeIngressContract;pub use realtime_ingress::RealtimeIngressError;pub use realtime_ingress::RealtimePayloadKind;pub use realtime_ingress::RealtimeTokenKind;pub use realtime_ingress::ValidatedRealtimeInput;pub use realtime_interpreter::complete_realtime_frame;pub use realtime_interpreter::prepare_realtime_frame;pub use realtime_interpreter::CompletedRealtimeFrame;pub use realtime_interpreter::PreparedRealtimeFrame;pub use realtime_interpreter::RealtimeFrameInterpretationError;pub use realtime_interpreter::RealtimeFrameTensorMechanisms;pub use realtime_mechanism_synthesis::synthesize_realtime_capabilities;pub use realtime_mechanism_synthesis::RealtimeMechanismFacts;pub use realtime_mechanism_synthesis::RealtimeMechanismSupport;pub use realtime_model::construct_realtime_model;pub use realtime_model::preflight_realtime_materialization_tasks;pub use realtime_model::realtime_task_binding_plan;pub use realtime_model::ConstructedRealtimeExecution;pub use realtime_model::ConstructedRealtimeModel;pub use realtime_model::PreparedRealtimeModelContract;pub use realtime_model::RealizedRealtimePolicy;pub use realtime_model::RealizedRealtimeState;pub use realtime_model::RealtimeArchitectureConstructionIdentity;pub use realtime_model::RealtimeArchitectureIdentity;pub use realtime_model::RealtimeLayerwiseRuntime;pub use realtime_model::RealtimeMaterializationComponent;pub use realtime_model::RealtimeMaterializationTask;pub use realtime_model::RealtimeModelConstructionError;pub use realtime_model::RealtimeModelConstructionMechanisms;pub use realtime_model::RealtimeModelContractError;pub use realtime_model::RealtimeTaskBindingPlan;pub use realtime_payload::RealtimePayloadContract;pub use realtime_payload::RealtimePayloadContractError;pub use realtime_payload::RealtimePayloadEnvelope;pub use realtime_payload::RealtimePayloadGeneration;pub use realtime_payload::RealtimePayloadHistory;pub use realtime_payload::RealtimePayloadHistoryError;pub use realtime_payload::RealtimePayloadOwnerIdentity;pub use realtime_payload_state::RealtimePayloadBranch;pub use realtime_payload_state::RealtimePayloadState;pub use realtime_payload_state::RealtimePayloadStateTransactionError;pub use realtime_selection::select_and_prepare_realtime_realization;pub use realtime_selection::select_realtime_realization;pub use realtime_selection::ConstructedRealtimeResources;pub use realtime_selection::PreparedRealtimeRealization;pub use realtime_selection::RealtimeArchitectureProof;pub use realtime_selection::RealtimeArchitectureRequirements;pub use realtime_selection::RealtimeContractError;pub use realtime_selection::RealtimeExecutionRequirements;pub use realtime_selection::RealtimeIdentity;pub use realtime_selection::RealtimeMechanism;pub use realtime_selection::RealtimeMechanismCapabilities;pub use realtime_selection::RealtimeMechanismRequirements;pub use realtime_selection::RealtimeObservationRequirements;pub use realtime_selection::RealtimePreparationError;pub use realtime_selection::RealtimeSelectionError;pub use realtime_selection::RealtimeSelectionIssue;pub use realtime_selection::RealtimeSelectionRequest;pub use realtime_selection::RealtimeTopologyPolicy;pub use realtime_selection::RealtimeWeightComponentRequirement;pub use realtime_selection::RealtimeWeightComponentRole;pub use realtime_selection::RealtimeWeightLoweringRequirement;pub use realtime_selection::SelectedRealtimeRealization;pub use realtime_selection::SelectedRealtimeStateComponentRealization;pub use realtime_selection::SelectedRealtimeStateRealization;pub use realtime_session::RealtimeHistoryGeneration;pub use realtime_session::RealtimeModelOwnerIdentity;pub use realtime_session::RealtimeModelSessionIdentity;pub use realtime_session::RealtimeSamplingReplacementError;pub use realtime_session::RealtimeSamplingUpdateError;pub use realtime_session::RealtimeSessionBranch;pub use realtime_session::RealtimeSessionError;pub use realtime_session::RealtimeSessionExecutionError;pub use realtime_session::RealtimeSessionIncarnation;pub use realtime_session::RealtimeSessionResumeError;pub use realtime_session::RealtimeSessionScheduler;pub use realtime_session::RealtimeSessionState;pub use realtime_session::RealtimeSessionTransactionError;pub use realtime_session::ReleasedRealtimeSession;pub use replicated_session::construct_replicated_text_session;pub use replicated_session::construct_replicated_text_session_with_execution;pub use replicated_session::construct_replicated_text_session_with_runtime;pub use replicated_session::prepare_default_partitioned_runtime;pub use replicated_session::prepare_layered_text_contract;pub use replicated_session::prepare_layered_text_contract_with_addressable_parameters;pub use replicated_session::prepare_partitioned_session_runtime;pub use replicated_session::prepare_partitioned_session_runtime_with_exclusions;pub use replicated_session::prepare_replicated_text_contract;pub use replicated_session::prepare_replicated_text_contract_with_addressable_parameters;pub use replicated_session::DirectReplicatedTextExecution;pub use replicated_session::DistributedSessionCheckpoint;pub use replicated_session::DistributedStateCheckpoint;pub use replicated_session::PartitionedRuntimeConstructionError;pub use replicated_session::PartitionedSessionFactoryInput;pub use replicated_session::PartitionedSessionPreparationError;pub use replicated_session::PartitionedUnitScope;pub use replicated_session::PredictionTargetOperation;pub use replicated_session::PreparedPartitionedRuntimeComponents;pub use replicated_session::PreparedPartitionedSessionRuntime;pub use replicated_session::PreparedReplicatedTextContract;pub use replicated_session::ReplicatedRuntimeExecutionStrategy;pub use replicated_session::ReplicatedTextExecutionStrategy;pub use replicated_session::ReplicatedTextRuntime;pub use replicated_session::ReplicatedTextSession;pub use replicated_session::ReplicatedTextSessionCheckpoint;pub use replicated_session::ReplicatedTextSessionError;pub use replicated_session::ReplicatedTextSessionMechanisms;pub use replicated_session::ReplicatedTextSessionReport;pub use replicated_session::RoutedReplicatedTextExecution;pub use replicated_session::SessionStateRealization;pub use replicated_session::TransactionalPromptCacheMechanisms;pub use replicated_text::group_replicated_text_transform_tasks;pub use replicated_text::locally_materialized_replicated_text_outputs;pub use replicated_text::partition_selected_replicated_text_materialization_tasks;pub use replicated_text::partitioned_replicated_text_materialization_tasks;pub use replicated_text::plan_local_replicated_text_materialization_tasks;pub use replicated_text::plan_replicated_text_materialization_tasks;pub use replicated_text::replicated_text_materialization_tasks;pub use replicated_text::select_replicated_text_realization;pub use replicated_text::selected_materialization_task_bytes;pub use replicated_text::AddressableStorageCapabilities;pub use replicated_text::AddressableStorageTiers;pub use replicated_text::BackendMechanismCapabilities;pub use replicated_text::GroupedOperationRequirement;pub use replicated_text::ParameterTransformConstraint;pub use replicated_text::ParameterTransformTarget;pub use replicated_text::ReplicatedTextArchitecture;pub use replicated_text::ReplicatedTextContractError;pub use replicated_text::ReplicatedTextMaterializationPartitionPlan;pub use replicated_text::ReplicatedTextMaterializationTask;pub use replicated_text::ReplicatedTextOutputCompanion;pub use replicated_text::ReplicatedTextOutputSelection;pub use replicated_text::ReplicatedTextParameterOwner;pub use replicated_text::ReplicatedTextParameterPresence;pub use replicated_text::ReplicatedTextParameterRequirement;pub use replicated_text::ReplicatedTextParameterRole;pub use replicated_text::ReplicatedTextPhysicalSource;pub use replicated_text::ReplicatedTextRequirements;pub use replicated_text::ReplicatedTextSelectionError;pub use replicated_text::ReplicatedTextSelectionRequest;pub use replicated_text::ReplicatedTextStateAccess;pub use replicated_text::ReplicatedTextTransformGroup;pub use replicated_text::SelectedParameterRealization;pub use replicated_text::SelectedReplicatedTextRealization;pub use replicated_text::SelectedStateComponentRealization;pub use replicated_text::SelectedStateRealization;pub use replicated_text::StateComponentMechanism;pub use replicated_text::StateComponentPlacement;pub use replicated_text::StateMechanismCapabilities;pub use replicated_text::StateStorageDtype;pub use replicated_text::WeightLoweringCapability;pub use replicated_text::WeightLoweringDescriptor;pub use replicated_text::WeightLoweringKind;pub use replicated_text::WeightResidencyMechanism;pub use residency::DeviceLayerWindow;pub use residency::OffloadUnit;pub use residency::QuantizationCompanionBindings;pub use residency::ResidencyAcquisition;pub use residency::ResidencyController;pub use residency::ResidencyControllerError;pub use residency::ResidencyDeclarationError;pub use residency::ResidencyLease;pub use residency::ResidencyLeaseOwner;pub use residency::ResidencyLeaseStorage;pub use residency::ResidencyReport;pub use residency::ResidencyTransfer;pub use residency::ResidencyTransferOwner;pub use residency::ResidencyWindowError;pub use residency::ResidencyWindowManager;pub use residency::ResidentLayerGroup;pub use residency::ResidentLayerGroupReport;pub use residency::WeightBinding;pub use residency::WeightBindingPlan;pub use residency::WeightBindingSelectionError;pub use residency::WeightMaterializationReport;pub use speculative::RunSpeculativeGeneration;pub use speculative::SpeculativeScheduler;pub use speculative_selection::select_and_prepare_speculative_realization;pub use speculative_selection::select_and_prepare_speculative_realization_observed;pub use speculative_selection::select_speculative_realization;pub use speculative_selection::ConstructedSpeculativeResources;pub use speculative_selection::PreparedSpeculativeRealization;pub use speculative_selection::SelectedSpeculativeCompletion;pub use speculative_selection::SelectedSpeculativePlacement;pub use speculative_selection::SelectedSpeculativeRealization;pub use speculative_selection::SelectedSpeculativeSampling;pub use speculative_selection::SelectedSpeculativeState;pub use speculative_selection::SpeculativeArchitectureCompatibilityProof;pub use speculative_selection::SpeculativeCaptureEntry;pub use speculative_selection::SpeculativeCaptureEnvelope;pub use speculative_selection::SpeculativeCaptureError;pub use speculative_selection::SpeculativeCaptureMetadata;pub use speculative_selection::SpeculativeCaptureSchema;pub use speculative_selection::SpeculativeContractError;pub use speculative_selection::SpeculativeIdentity;pub use speculative_selection::SpeculativeLaneIdentity;pub use speculative_selection::SpeculativeMechanism;pub use speculative_selection::SpeculativeMechanismCapabilities;pub use speculative_selection::SpeculativeMechanismRequirements;pub use speculative_selection::SpeculativePlacementRequest;pub use speculative_selection::SpeculativePreparationError;pub use speculative_selection::SpeculativeRealizationRequirements;pub use speculative_selection::SpeculativeSelectionError;pub use speculative_selection::SpeculativeSelectionRequest;pub use speculative_selection::SpeculativeStateCacheIdentityIngredients;pub use speculative_selection::SpeculativeStrategyClass;pub use speculative_selection::SpeculativeStrategyRequirements;pub use state::realize_architecture_state;pub use state::ArchitectureStateFactory;pub use state::ArchitectureStatePartitionError;pub use state::ArchitectureStatePartitionPlan;pub use state::ArchitectureStatePartitionRule;pub use state::ArchitectureStatePlacement;pub use state::ArchitectureStateRealizationError;pub use state::DeviceState;pub use state::LayerRuntimeState;pub use state::ModelStateIdentity;pub use state::ResettableRuntimeLayerState;pub use state::ResettableRuntimeState;pub use state::RuntimeLayerState;pub use state::RuntimeState;pub use state::RuntimeStateComponents;pub use state::StateError;pub use state::StateLayout;pub use state::StateSegmentId;pub use state::StateSegmentLifetime;pub use state::StateSegmentSpec;pub use state::DEFAULT_STATE_SEGMENT_ID;
Modules§
- automatic_
support - Portable automatic-plan resource sizing and telemetry projections. Portable resource sizing and telemetry for automatic execution planning.
- backend
- Backend execution, parameter, transfer, and collective capabilities. Narrow capability contracts implemented by execution backends.
- cache
- Backend-neutral mutable-cache ownership, storage, and admission algorithms. Backend-neutral ownership and admission algorithms for live model caches.
- communication
- Opaque groups, routes, and capability contracts for distributed mechanisms. Opaque communication manifests projected from semantic parallel topology.
- component
- Typed multimodal component graphs and residency accounting. Typed multimodal component graphs and residency classes.
- composite
- Mechanism-only selection contracts for composite model input. Mechanism-only selection contracts for composite model input.
- decision
- Backend-neutral sequential prediction decisions and layered handoff. Backend-neutral sequential prediction decisions and layered traversal handoff.
- dense
- Backend-neutral dense-stream residency telemetry. Backend-neutral dense-stream residency telemetry.
- draft
- Backend-neutral speculative-state fork, commit, and rollback ownership. Transactional ownership for embedded draft mutable state.
- execution
- Portable execution-group topology and scheduling state. Validated execution-group dependency graphs and ready-set scheduling.
- execution_
plan - Portable execution-plan policy derivation before backend resource realization.
- expert
- Runtime ownership boundary for routed expert acquisition and residency.
- generation
- Backend-neutral causal-model and token-sampling contracts. Backend-neutral causal-model and token-sampling contracts.
- input
- Backend-neutral ownership of prepared multimodal tensors. Backend-neutral ownership of prepared multimodal tensors.
- inspection
- Backend-neutral activation, target-state, and routed-expert observation contracts.
- layered
- Statically dispatched layered architecture lifecycle and resident policy. Statically dispatched layered-architecture lifecycle and resident execution.
- load_
request - Normalized portable policy for cold model preparation. Normalized portable policy for cold model preparation.
- mechanism_
synthesis - Exact mechanism capability synthesis from neutral requirements. Requirement-driven synthesis of exact backend mechanism capabilities.
- parallel
- Architecture-declared parallel parameter semantics and local layouts. Backend-neutral semantic parameter sharding and rank-local layouts.
- parameter
- Neutral checkpoint materialization and stable parameter binding. Backend-neutral checkpoint materialization and parameter binding.
- partition
- Backend-neutral rank-local architecture ownership. Backend-neutral ownership of one rank-local architecture partition.
- partitioned_
execution - Rank-local graph execution over opaque communication resources. Backend-neutral rank-local execution over opaque communication resources.
- placement
- Neutral lowering of logical parameter placement into bounded checkpoint selections.
- prefetch
- Backend-neutral bounded background weight-prefetch execution. Backend-neutral bounded background weight prefetch execution.
- realtime
- Atomic realtime model, schedule, sampler, and random-state transactions. Atomic runtime ownership for one realtime model-generation transition.
- realtime_
executor - Complete family-blind realtime frame coordination. Mandatory family-blind coordination of one complete realtime frame.
- realtime_
ingress - Portable realtime input validation before opaque token materialization. Portable realtime-frame validation before opaque token materialization.
- realtime_
interpreter - Neutral delayed-frame interpretation over opaque token mechanisms. Family-blind interpretation of portable realtime schedule transitions.
- realtime_
mechanism_ synthesis - Requirement enumeration for realtime mechanisms, independent of native resources.
- realtime_
model - Family-blind construction of selected layered realtime models. Family-blind construction of a selected layered realtime model.
- realtime_
payload - Backend-neutral payload retention for delayed realtime coordinates. Backend-neutral payload retention for delayed realtime coordinates.
- realtime_
payload_ state - Atomic model-state and delayed-payload-history transactions. Atomic model-state and delayed-payload-history transactions.
- realtime_
selection - Family-blind realtime requirements, selection, and construction gating. Family-blind realtime requirements, selection, and construction gating.
- realtime_
session - Backend-neutral realtime session ownership over the singular fair scheduler. Backend-neutral realtime session ownership over the singular fair scheduler.
- replicated_
session - Backend-neutral replicated-text execution and session ownership. Backend-neutral replicated-text execution and session ownership.
- replicated_
text - Selection contracts for replicated text architectures. Selection contracts for replicated text architectures.
- residency
- Backend-neutral immutable-weight residency declarations and orchestration. Backend-neutral immutable-weight residency declarations and control state.
- speculative
- Backend-neutral speculative request lifecycle and fair scheduling. Backend-neutral speculative request lifecycle and fair scheduling.
- speculative_
selection - Family-blind speculative requirements, selection, and construction gating. Family-blind selection and construction gates for speculative execution.
- state
- Architecture-declared mutable state and concrete runtime realizations. Backend-neutral mutable model-state contracts.
Structs§
- Dense
Disk Stream Load Options - Controls for bounded dense-unit streaming from checkpoint storage.
- Dense
Transfer Schedule - Ordered bounded transfer cursor used by dense streamed execution.
- Layerwise
Load Options - Loader controls for a host-backed layerwise execution engine.
- Layerwise
Model Metadata - Inspectable backend-neutral parameter-residency metadata for a layered model.
- Parameter
Bank Key - Stable mechanism identity for one member of an independently addressable bank.
- Parameter
Bank Load Options - Backend-neutral controls for independently addressable parameter-bank residency.
- Static
Unit Bindings - One pinned static module and its checkpoint bindings.
Enums§
- Dense
Transfer Schedule Error - Invalid transition in a bounded dense transfer schedule.
- Execution
Residency - Static-parameter placement used by the generalized execution engine.
- Expert
Pass - Execution-path classification for routed-expert telemetry and chunking.
- Layer
Weight Residency - Placement policy for ordinary architecture execution units.
- Ordinary
Weight Residency - Placement of ordinary parameters beside independently addressable banks.
- Parameter
Bank Access - Generic workload class for independently addressable storage access.
- Parameter
Bank Residency - Independently addressable bank placement relative to ordinary execution units.
- Weight
Residency - Composable ordinary-unit and independently addressable bank placement.
- Weight
Residency Policy Error - Invalid backend-neutral immutable-weight residency policy.
Constants§
- DENSE_
TRANSFER_ WINDOW - Current plus next unit retained by dense streamed execution.