SAP AI SDK for JavaScript - v2.13.0
    Preparing search index...

    Type Alias OrchestrationCallOptions

    OrchestrationCallOptions: Pick<
        BaseChatModelCallOptions,
        | "stop"
        | "signal"
        | "timeout"
        | "callbacks"
        | "metadata"
        | "runId"
        | "runName"
        | "tags"
        | "ls_structured_output_format",
    > & {
        cache_control?: CacheControl;
        customRequestConfig?: CustomRequestConfig;
        placeholderValues?: Record<string, string>;
        promptIndex?: number;
        responseFormat?:
            | ResponseFormatText
            | ResponseFormatJsonObject
            | ResponseFormatJsonSchema;
        streamOptions?: StreamOptions;
        strict?: boolean;
        tools?: ChatOrchestrationToolType[];
    }

    Options for an orchestration call.

    Type Declaration

    • Optionalcache_control?: CacheControl

      Cache control configuration for prompt caching. When provided, a cache breakpoint is automatically applied to the last cacheable text block of the last message so the breakpoint advances naturally as the conversation grows. This removes the need to place cache_control on individual content blocks manually.

      Only applies to models that support cache_control through orchestration (Anthropic Claude and Amazon Nova model families). Other models will ignore the directive. See the SAP AI Core prompt caching docs for supported models and breakpoint limits.

    • OptionalcustomRequestConfig?: CustomRequestConfig
    • OptionalplaceholderValues?: Record<string, string>
    • OptionalpromptIndex?: number
    • OptionalresponseFormat?: ResponseFormatText | ResponseFormatJsonObject | ResponseFormatJsonSchema
    • OptionalstreamOptions?: StreamOptions
    • Optionalstrict?: boolean
    • Optionaltools?: ChatOrchestrationToolType[]

    Settings
    Member Visibility