VisionAccelerationSettings

Settings for VisionAcceleration. Values fall back through three levels: document → SDK → built-in default. Writes target the document only when set on a document’s settings, otherwise the SDK globally when set on SdkSettings.

Tags: Vision, Advanced

from nutrient_sdk import VisionAccelerationSettings

Construction

# Per-document override
with Document.open("input.pdf") as doc:
    settings = doc.settings.vision_acceleration_settings
    settings.some_field = new_value          # mutate fields directly

# SDK-wide default (applies to all documents)
SdkSettings.vision_acceleration_settings.some_field = new_value

Settings are configured by writing to fields on the returned object. The settings property itself cannot be reassigned — doc.settings.vision_acceleration_settings = other_settings is rejected.

Properties

gpu_excluded_modules

@property
def gpu_excluded_modules(self) -> list

@gpu_excluded_modules.setter
def gpu_excluded_modules(self, value: list) -> None

Internal denylist of Vision capacity names that must run on the CPU even when Provider would otherwise put them on the GPU. Holds the models that don’t (yet) work on the WebGPU execution provider; remove a name once its GPU path is fixed and validated. Hidden from the public settings schema. The values are capacity Names (e.g. "InferencePicture").

Type: list

Default: Array.Empty<string>()

provider

@property
def provider(self) -> ExecutionProvider

@provider.setter
def provider(self, value: ExecutionProvider) -> None

Execution provider for Vision ONNX model inference. Determines whether GPU-eligible models are offloaded to the GPU. Changing this at runtime (e.g. via the daemon) takes effect the next time a model session is (re)built.

Type: ExecutionProvider

Default: ExecutionProvider.Cpu