ONNX Runtime JavaScript API
    Preparing search index...

    Interface CoreMLExecutionProviderOption

    interface CoreMLExecutionProviderOption {
        coreMlFlags?: number;
        enableOnSubgraph?: boolean;
        name: "coreml";
        onlyEnableDeviceWithANE?: boolean;
        useCPUAndGPU?: boolean;
        useCPUOnly?: boolean;
    }

    Hierarchy (View Summary)

    Index

    Properties

    coreMlFlags?: number

    The bit flags for CoreML execution provider.

    COREML_FLAG_USE_CPU_ONLY = 0x001
    COREML_FLAG_ENABLE_ON_SUBGRAPH = 0x002
    COREML_FLAG_ONLY_ENABLE_DEVICE_WITH_ANE = 0x004
    COREML_FLAG_ONLY_ALLOW_STATIC_INPUT_SHAPES = 0x008
    COREML_FLAG_CREATE_MLPROGRAM = 0x010
    COREML_FLAG_USE_CPU_AND_GPU = 0x020

    See include/onnxruntime/core/providers/coreml/coreml_provider_factory.h for more details.

    This flag is available only in ONNXRuntime (Node.js binding).

    enableOnSubgraph?: boolean

    Specify whether to enable CoreML EP on subgraph.

    This setting is available only in ONNXRuntime (react-native).

    name: "coreml"
    onlyEnableDeviceWithANE?: boolean

    Specify whether to only enable CoreML EP for Apple devices with ANE (Apple Neural Engine).

    This setting is available only in ONNXRuntime (react-native).

    useCPUAndGPU?: boolean
    useCPUOnly?: boolean

    Specify whether to use CPU only in CoreML EP.

    This setting is available only in ONNXRuntime (react-native).