Extract text from a document using vision language models (VLMs). VLMs can understand document layout and structure more intelligently than traditional OCR.
| Name | Type | Required | Description |
|---|---|---|---|
| llm_spec | LlmSpec | Yes | |
| preprocessing_configuration | Optional[VlmPreprocessingConfig] | No | |
| image_spec | Optional[ImageSpec] | No | |
| output_format | TextOutputFormat | Yes | |
| page_range | Optional[PageRange] | No | |
| type | Literal["extractVlmText"] | Yes | None |