pyflink.dataframe.ai.llm.LLMAccessor.ai_extract

LLMAccessor.ai_extract(input_col: str | Expression, schema: str, *, provider: str = None, model: str = None, config: Dict[str, str] = None) → DataFrame

Extract structured information from text.

Parameters:

• input_col – Column name (str) or Expression for the input text.

• schema – JSON schema string describing the fields to extract, e.g. '{"name":"STRING", "phone":"STRING"}'.

• provider – Provider name.

• model – Model name.

• config – Optional runtime config.

Returns:

• extracted_json (STRING): extracted fields as a JSON string.

Return type:

A new DataFrame with a column appended

Example:

>>> df.llm.ai_extract("text",
...     '{"name":"STRING", "phone":"STRING"}', model="qwen-plus")