AI Functions

These dataclasses are used in the SDK to represent API requests and responses for services in the databricks.sdk.service.aifunctions module.

class databricks.sdk.service.aifunctions.AiClassifyOptions(enable_confidence_scores: bool | None = None, enable_rationales: bool | None = None, instructions: str | None = None, multilabel: bool | None = None, version: str | None = None)

ai_classify

enable_confidence_scores: bool | None = None

When true, includes a per-label confidence score in the response.

enable_rationales: bool | None = None

When true, includes a rationale explaining each classification in the response.

instructions: str | None = None

Natural-language guidance that steers how the text is classified (up to 20,000 characters).

multilabel: bool | None = None

When true, allows more than one label to be returned per input.

version: str | None = None

The function version to invoke. Defaults to the latest version. Supported versions: [“2.1”].

as_dict() → dict

Serializes the AiClassifyOptions into a dictionary suitable for use as a JSON request body.

as_shallow_dict() → dict

Serializes the AiClassifyOptions into a shallow dictionary of its immediate attributes.

classmethod from_dict(d: Dict[str, Any]) → AiClassifyOptions

Deserializes the AiClassifyOptions from a dictionary.

class databricks.sdk.service.aifunctions.AiClassifyResponse(metadata: 'Optional[AiClassifyResponseMetadata]' = None, response: 'Optional[any]' = None)

metadata: AiClassifyResponseMetadata | None = None

Additional metadata returned by AI Classify.

response: any | None = None

The function result as a JSON value. An array of per-label objects: one element in single-label mode (the default), or multiple elements when multilabel is true. When enable_confidence_scores and enable_rationales are true, confidence_score and rationale are included in each response value, respectively.

as_dict() → dict

Serializes the AiClassifyResponse into a dictionary suitable for use as a JSON request body.

as_shallow_dict() → dict

Serializes the AiClassifyResponse into a shallow dictionary of its immediate attributes.

classmethod from_dict(d: Dict[str, Any]) → AiClassifyResponse

Deserializes the AiClassifyResponse from a dictionary.

class databricks.sdk.service.aifunctions.AiClassifyResponseMetadata(version: 'Optional[str]' = None)

version: str | None = None

The resolved function version.

as_dict() → dict

Serializes the AiClassifyResponseMetadata into a dictionary suitable for use as a JSON request body.

as_shallow_dict() → dict

Serializes the AiClassifyResponseMetadata into a shallow dictionary of its immediate attributes.

classmethod from_dict(d: Dict[str, Any]) → AiClassifyResponseMetadata

Deserializes the AiClassifyResponseMetadata from a dictionary.

class databricks.sdk.service.aifunctions.AiExtractBbox(coord: List[int] | None = None, page_id: int | None = None)

A bounding box on a source page; used by bbox-input citations.

coord: List[int] | None = None

Pixel coordinates on the page image as [x0, y0, x1, y1].

page_id: int | None = None

0-based page index the box is on.

as_dict() → dict

Serializes the AiExtractBbox into a dictionary suitable for use as a JSON request body.

as_shallow_dict() → dict

Serializes the AiExtractBbox into a shallow dictionary of its immediate attributes.

classmethod from_dict(d: Dict[str, Any]) → AiExtractBbox

Deserializes the AiExtractBbox from a dictionary.

class databricks.sdk.service.aifunctions.AiExtractCitation(bbox: List[AiExtractBbox] | None = None, id: int | None = None, start: int | None = None, stop: int | None = None)

A citation locating an extracted value in the source. start/stop are set for span (STRING input) citations, bbox for bbox (parsed-document input) citations.

bbox: List[AiExtractBbox] | None = None

Bounding boxes locating the citation on the source pages; set for bbox citations.

id: int | None = None

Integer matching a citation_ids entry on an extracted field.

start: int | None = None

Inclusive 0-based character offset into the input string; set for span citations.

stop: int | None = None

Exclusive 0-based character offset into the input string; set for span citations.

as_dict() → dict

Serializes the AiExtractCitation into a dictionary suitable for use as a JSON request body.

as_shallow_dict() → dict

Serializes the AiExtractCitation into a shallow dictionary of its immediate attributes.

classmethod from_dict(d: Dict[str, Any]) → AiExtractCitation

Deserializes the AiExtractCitation from a dictionary.

class databricks.sdk.service.aifunctions.AiExtractOptions(enable_citations: bool | None = None, enable_confidence_scores: bool | None = None, instructions: str | None = None, mode: str | None = None, version: str | None = None)

ai_extract

enable_citations: bool | None = None

When true, includes citation metadata locating each extracted value in the source. Depending on the type of input, citations can be one of two types:

For raw text (STRING) inputs, a citation is a span of text in the original input. Each object in metadata.citations has an id (integer matching a citation_ids entry on a field), a start (inclusive 0-based character offset into the input string), and a stop (exclusive 0-based character offset into the input string).

For PDF documents and images (when using ai_extract downstream of ai_parse_document), a citation is a bounding box in the original input. Each object in metadata.citations has an id (integer matching a citation_ids entry on a field) and a bbox (array of {coord, page_id} objects, identical in shape to element.bbox in ai_parse_document output; coord is pixel coordinates on the page image as [x0, y0, x1, y1], and page_id is a 0-based page index).

enable_confidence_scores: bool | None = None

When true, includes a per-field confidence score in the response.

instructions: str | None = None

Natural-language guidance that steers how data is extracted (up to 20,000 characters).

mode: str | None = None

Extraction mode. Supported modes: “precision” — more powerful extraction for complex schemas, long documents, and reasoning-heavy extractions. Defaults to none (standard extraction).

version: str | None = None

The function version to invoke. Defaults to the latest version. Supported versions: [“2.1”].

as_dict() → dict

Serializes the AiExtractOptions into a dictionary suitable for use as a JSON request body.

as_shallow_dict() → dict

Serializes the AiExtractOptions into a shallow dictionary of its immediate attributes.

classmethod from_dict(d: Dict[str, Any]) → AiExtractOptions

Deserializes the AiExtractOptions from a dictionary.

class databricks.sdk.service.aifunctions.AiExtractResponse(metadata: 'Optional[AiExtractResponseMetadata]' = None, response: 'Optional[any]' = None)

metadata: AiExtractResponseMetadata | None = None

Additional metadata returned by AI Extract.

response: any | None = None

The function result as a JSON value. When enable_confidence_scores and enable_citations are true, confidence and citation_ids are included in each response field, respectively.

as_dict() → dict

Serializes the AiExtractResponse into a dictionary suitable for use as a JSON request body.

as_shallow_dict() → dict

Serializes the AiExtractResponse into a shallow dictionary of its immediate attributes.

classmethod from_dict(d: Dict[str, Any]) → AiExtractResponse

Deserializes the AiExtractResponse from a dictionary.

class databricks.sdk.service.aifunctions.AiExtractResponseMetadata(chunk_type: 'Optional[str]' = None, citations: 'Optional[List[AiExtractCitation]]' = None, mode: 'Optional[str]' = None, version: 'Optional[str]' = None)

chunk_type: str | None = None

How the source was chunked for citation offsets (span for text input, bbox for parsed-document input); present when citations are enabled.

citations: List[AiExtractCitation] | None = None

Citation objects locating each result in the source; present when citations are enabled.

mode: str | None = None

The resolved extraction mode; present when a non-default mode was used.

version: str | None = None

The resolved function version.

as_dict() → dict

Serializes the AiExtractResponseMetadata into a dictionary suitable for use as a JSON request body.

as_shallow_dict() → dict

Serializes the AiExtractResponseMetadata into a shallow dictionary of its immediate attributes.

classmethod from_dict(d: Dict[str, Any]) → AiExtractResponseMetadata

Deserializes the AiExtractResponseMetadata from a dictionary.

class databricks.sdk.service.aifunctions.AiParseDocumentFileMetadata(file_modification_time: str | None = None, file_name: str | None = None, file_path: str | None = None, file_size: int | None = None)

Metadata about the source file; present only for file-path input.

file_modification_time: str | None = None

Last-modified timestamp of the source file, as an HTTP date string.

file_name: str | None = None

Base name of the source file.

file_path: str | None = None

Unity Catalog volume path of the source file.

file_size: int | None = None

Size of the source file in bytes.

as_dict() → dict

Serializes the AiParseDocumentFileMetadata into a dictionary suitable for use as a JSON request body.

as_shallow_dict() → dict

Serializes the AiParseDocumentFileMetadata into a shallow dictionary of its immediate attributes.

classmethod from_dict(d: Dict[str, Any]) → AiParseDocumentFileMetadata

Deserializes the AiParseDocumentFileMetadata from a dictionary.

class databricks.sdk.service.aifunctions.AiParseDocumentOptions(description_element_types: str | None = None, image_output_path: str | None = None, page_range: str | None = None, version: str | None = None)

ai_parse_document

description_element_types: str | None = None

Element types for which an AI-generated description is produced. Use “*” (default) to generate descriptions for all supported element types, “figure” to generate them for figures only, or “” (empty string) to generate none. Only figure descriptions are supported for version “2.0”, so “*” and “figure” produce the same behavior.

image_output_path: str | None = None

Unity Catalog volume path where rendered page and element images are written.

page_range: str | None = None

Pages to parse (1-indexed), as a comma-separated list of page numbers or ranges (e.g. “1,3,5-10”).

version: str | None = None

The ai_parse_document output schema version. Supported value: “2.0”.

as_dict() → dict

Serializes the AiParseDocumentOptions into a dictionary suitable for use as a JSON request body.

as_shallow_dict() → dict

Serializes the AiParseDocumentOptions into a shallow dictionary of its immediate attributes.

classmethod from_dict(d: Dict[str, Any]) → AiParseDocumentOptions

Deserializes the AiParseDocumentOptions from a dictionary.

class databricks.sdk.service.aifunctions.AiParseDocumentPageError(error_message: str | None = None, page_id: int | None = None)

A single page that failed to parse while the overall request succeeded.

error_message: str | None = None

Message describing why the page failed.

page_id: int | None = None

0-based index of the page that failed.

as_dict() → dict

Serializes the AiParseDocumentPageError into a dictionary suitable for use as a JSON request body.

as_shallow_dict() → dict

Serializes the AiParseDocumentPageError into a shallow dictionary of its immediate attributes.

classmethod from_dict(d: Dict[str, Any]) → AiParseDocumentPageError

Deserializes the AiParseDocumentPageError from a dictionary.

class databricks.sdk.service.aifunctions.AiParseDocumentResponse(document: 'Optional[any]' = None, error_status: 'Optional[List[AiParseDocumentPageError]]' = None, metadata: 'Optional[AiParseDocumentResponseMetadata]' = None)

document: any | None = None

The parsed document as a JSON value, containing the extracted pages and elements.

error_status: List[AiParseDocumentPageError] | None = None

Per-page partial-failure details; present when the request succeeds (2xx) but individual pages fail.

metadata: AiParseDocumentResponseMetadata | None = None

Additional metadata returned by AI Parse Document.

as_dict() → dict

Serializes the AiParseDocumentResponse into a dictionary suitable for use as a JSON request body.

as_shallow_dict() → dict

Serializes the AiParseDocumentResponse into a shallow dictionary of its immediate attributes.

classmethod from_dict(d: Dict[str, Any]) → AiParseDocumentResponse

Deserializes the AiParseDocumentResponse from a dictionary.

class databricks.sdk.service.aifunctions.AiParseDocumentResponseMetadata(file_metadata: 'Optional[AiParseDocumentFileMetadata]' = None, id: 'Optional[str]' = None, version: 'Optional[str]' = None)

file_metadata: AiParseDocumentFileMetadata | None = None

Describes the source file; present only for file-path input.

id: str | None = None

Unique identifier for the parse request.

version: str | None = None

The resolved function version.

as_dict() → dict

Serializes the AiParseDocumentResponseMetadata into a dictionary suitable for use as a JSON request body.

as_shallow_dict() → dict

Serializes the AiParseDocumentResponseMetadata into a shallow dictionary of its immediate attributes.

classmethod from_dict(d: Dict[str, Any]) → AiParseDocumentResponseMetadata

Deserializes the AiParseDocumentResponseMetadata from a dictionary.