Nutrient Java SDK/io.nutrient.sdk.settings/AiProcessingSettings

AiProcessingSettings

Settings for AiProcessing. 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.

Functions

getAiProvider

public final VlmProvider getAiProvider()

The primary AI provider for extraction/classification.

getApiKey

public final String getApiKey()

API key for the configured provider.

getEndpoint

public final String getEndpoint()

Endpoint URL.

getFallbackAiProvider

public final VlmProvider getFallbackAiProvider()

Optional secondary AI provider used to transparently retry a request when the primary returns no content (content filtering) or is temporarily unavailable.

getIncludeCompositeConfidence

public final boolean getIncludeCompositeConfidence()

(Schema v1 only.) Daemon-level default for whether the legacy rolled-up composite number is attached to result.json metadata leaves as confidence.

getIncludeConfidence

public final boolean getIncludeConfidence()

Daemon-level default for whether AI processing capacities emit confidence output - the confidence.json sidecar and the per-field confidence signals on result.json metadata.

getIncludeGroundedness

public final boolean getIncludeGroundedness()

Daemon-level default for whether the NLI groundedness score is computed and surfaced as the primary confidence.groundedness (schema v2 only).

getIncludePageImages

public final boolean getIncludePageImages()

Opt-in (default ): when true, every page of the source document is rendered to an image and sent to the model alongside the IR-lite text (multimodal extraction).

getIncludeSourceLocations

public final boolean getIncludeSourceLocations()

Whether extraction grounds each extracted field back to its source location in the document — the metadata node of result.json with per-field match labels and source bounding boxes.

getMaxAttempts

public final Integer getMaxAttempts()

Maximum retry attempts on structured-output failure.

getMaxInputTokens

public final int getMaxInputTokens()

Input-token budget per model call.

getMaxParallelCalls

public final int getMaxParallelCalls()

Maximum number of windowed extraction calls to run concurrently.

getMetadataSchemaVersion

public final int getMetadataSchemaVersion()

Daemon-level default for the result.json metadata schema version: 1 = the legacy flat shape (per-field match / confidenceComponents / opt-in composite confidence); 2 = the simplified citation + confidence shape (primary groundedness, ocr, boolean schemaValid).

getModel

public final String getModel()

Model identifier (e.g.

getProvider

public final String getProvider()

Provider discriminator.

getStrictStructuredOutput

public final boolean getStrictStructuredOutput()

Opt-in (default ): when true, structured output runs in the provider's strict mode — the response is grammar-constrained to the schema, which is normalized automatically (additionalProperties: false, all properties required, unsupported keywords moved into descriptions).

getTemperature

public final Double getTemperature()

Sampling temperature passed to the model.

getUnparseableOutputHandling

public final UnparseableOutputHandling getUnparseableOutputHandling()

How extraction responds when the model's output never parses against the requested schema.

setAiProvider

public final void setAiProvider(VlmProvider value)

The primary AI provider for extraction/classification.

setApiKey

public final void setApiKey(String value)

API key for the configured provider.

setEndpoint

public final void setEndpoint(String value)

Endpoint URL.

setFallbackAiProvider

public final void setFallbackAiProvider(VlmProvider value)

Optional secondary AI provider used to transparently retry a request when the primary returns no content (content filtering) or is temporarily unavailable.

setIncludeCompositeConfidence

public final void setIncludeCompositeConfidence(boolean value)

(Schema v1 only.) Daemon-level default for whether the legacy rolled-up composite number is attached to result.json metadata leaves as confidence.

setIncludeConfidence

public final void setIncludeConfidence(boolean value)

Daemon-level default for whether AI processing capacities emit confidence output - the confidence.json sidecar and the per-field confidence signals on result.json metadata.

setIncludeGroundedness

public final void setIncludeGroundedness(boolean value)

Daemon-level default for whether the NLI groundedness score is computed and surfaced as the primary confidence.groundedness (schema v2 only).

setIncludePageImages

public final void setIncludePageImages(boolean value)

Opt-in (default ): when true, every page of the source document is rendered to an image and sent to the model alongside the IR-lite text (multimodal extraction).

setIncludeSourceLocations

public final void setIncludeSourceLocations(boolean value)

Whether extraction grounds each extracted field back to its source location in the document — the metadata node of result.json with per-field match labels and source bounding boxes.

setMaxAttempts