async-bulkhead-llm - v3.17.0
    Preparing search index...

    Type Alias LLMRunContext

    Context passed to run() callbacks for mid-flight usage reporting.

    type LLMRunContext = {
        admission: LLMRunAdmission;
        admissionClass?: string;
        admissionId: string;
        bypassDetail?: LLMRejectDetail;
        bypassReason?: LLMShadowableRejectReason;
        limitRevision: number;
        reservation: LLMReservationEstimate | null;
        resources: LLMAdmissionResources;
        abandonBorrowedConcurrency(): boolean;
        reportUsage(
            usage: TokenUsage,
            reconciliation?: ProgressiveReconciliationOptions,
        ): UsageReport;
    }
    Index
    admission admissionClass? admissionId bypassDetail? bypassReason? limitRevision reservation resources
    abandonBorrowedConcurrency reportUsage
    admission: LLMRunAdmission

    Whether the callback holds real bulkhead capacity.

    admissionClass?: string

    Bounded policy class used for this execution, when configured.

    admissionId: string

    Stable identifier for this execution. Bypasses use a shadow- prefix.

    bypassDetail?: LLMRejectDetail

    Capacity snapshot associated with the bypass decision, when available.

    bypassReason?: LLMShadowableRejectReason

    Simulated or authoritative reason when admission === "bypassed".

    limitRevision: number

    Immutable limit revision associated with this admission or bypass decision.

    reservation: LLMReservationEstimate | null

    Exact reservation evaluated for this call, or null without a budget.

    resources: LLMAdmissionResources

    Exact protected/shared resource attribution for this execution.

    • Return a borrowed local concurrency slot before final settlement while retaining any token hold. Returns true only for the first effective call.

      Returns boolean

    Settings
    Member Visibility

    On This Page
    Properties
    Methods