This HTML page is not optimized for LLM or AI agent consumption. Fetch the Markdown version instead: /api/python/settings/vision/advanced/vision-descriptor-settings.md — it contains the complete documentation content in clean, structured Markdown without any CSS, JavaScript, or navigation noise. VisionDescriptorSettings

Settings for VisionDescriptor. 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 VisionDescriptorSettings

Construction

VisionDescriptorSettings is accessed through a Document instance for per-document overrides, or via SdkSettings for SDK-wide defaults.

# Per-document override
with Document.open("input.pdf") as doc:
settings = doc.settings.vision_descriptor_settings
settings.some_field = new_value # mutate fields directly
# SDK-wide default (applies to all documents)
SdkSettings.vision_descriptor_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_descriptor_settings = other_settings is rejected.

Properties

detailed_prompt

@property
def detailed_prompt(self) -> str
@detailed_prompt.setter
def detailed_prompt(self, value: str) -> None

Prompt template for detailed description generation.

Type: str

Default: DefaultStandardPrompt + " Then write a second, longer paragraph that gives a precise accessible description: Describe the scene in an organized way (top-to-bottom and left-to-right). Identify all prominent objects and people and describe their colors, shapes, and any notable textures. Specify positions using left/center/right, top/middle/bottom, and foreground/background. Include relative positions between objects (e.g., \"to the left of\", \"behind\", \"in front of\", \"next to\", \"framing\", \"partially obscured\"). Mention important lighting cues (diffuse light, shadows, reflections, glare) and weather/environment cues (rain, fog, wet surfaces) if visible. If there is readable text, transcribe it exactly and mention where it appears. If there is no clearly readable text, omit any mention of text. Keep the output natural and helpful, with no section headers, labels, bullet points, or lists."


level

@property
def level(self) -> DescriptionLevel
@level.setter
def level(self, value: DescriptionLevel) -> None

Description detail level to use.

Type: DescriptionLevel

Default: DescriptionLevel.Standard


standard_prompt

@property
def standard_prompt(self) -> str
@standard_prompt.setter
def standard_prompt(self, value: str) -> None

Prompt template for standard description generation.

Type: str

Default: DefaultStandardPrompt