Configuration Types

Configuration options for Reducto API operations

Settings

General document processing settings.

document_password · string

Password to decrypt password-protected documents.

embed_pdf_metadata · boolean

If True, embed OCR metadata into the returned PDF. Defaults to False.

extraction_mode · enum

The mode to use for text extraction from PDFs. One of:

• ocr - Uses optical character recognition only
• hybrid - Combines OCR with embedded PDF text for best accuracy (default)

force_file_extension · string

Force the URL to be downloaded as a specific file extension (e.g. .png).

force_url_result · boolean

Force the result to be returned in URL form.

ocr_system · enum

OCR system to use. One of:

• standard - Best multilingual OCR system
• legacy - Only supports Germanic languages (for backwards compatibility)

page_range · Union[PageRange, List[PageRange], List[int]]

The page range to process (1-indexed). By default, the entire document is processed.

persist_results · boolean

If True, persist the results indefinitely. Defaults to False.

return_images · List[enum]

Whether to return images for the specified block types. Options: figure, table. By default, no images are returned.

return_ocr_data · boolean

If True, return OCR data in the result. Defaults to False.

timeout · float

The timeout for the job in seconds.

Enhance

Configuration for enhancing extraction accuracy using AI models.

agentic · List[Union[TableAgentic, FigureAgentic, TextAgentic]]

Agentic uses vision language models to enhance the accuracy of the output of different types of extraction. This will incur a cost and latency increase.

summarize_figures · boolean

If True, summarize figures using a small vision language model. Defaults to True.

TableAgentic

scope · literal · required

Always set to “table”.

prompt · string

Custom prompt for table agentic processing.

FigureAgentic

scope · literal · required

Always set to “figure”.

advanced_chart_agent · boolean

If True, use the advanced chart agent. Defaults to False.

prompt · string

Custom prompt for figure agentic processing.

return_overlays · boolean

If True, return overlays for the figure. This allows you to verify the quality of the extraction.

TextAgentic

scope · literal · required

Always set to “text”.

prompt · string

Custom instructions for agentic text. Note: This only applies to form regions (key-value).

Formatting

Configuration for output formatting options.

add_page_markers · boolean

If True, add page markers to the output. Defaults to False. Useful for extracting data with page specific information.

include · List[enum]

A list of formatting to include in the output. Options:

• change_tracking
• highlight
• comments
• hyperlinks
• signatures

merge_tables · boolean

A flag to indicate if consecutive tables with the same number of columns should be merged. Defaults to False.

table_output_format · enum

The mode to use for table output. Options:

• html - HTML table format
• json - JSON array format
• md - Markdown table format
• jsonbbox - JSON with bounding boxes
• dynamic - Returns md for simpler tables and html for complex tables (default)
• csv - CSV format

Retrieval

Configuration for retrieval and chunking behavior.

chunking · Chunking

Chunking configuration for the document.

embedding_optimized · boolean

If True, use embedding optimized mode. Defaults to False.

filter_blocks · List[enum]

A list of block types to filter out from ‘content’ and ‘embed’ fields. By default, no blocks are filtered. Options:

• Header
• Footer
• Title
• Section Header
• Page Number
• List Item
• Figure
• Table
• Key Value
• Text
• Comment
• Signature

Chunking

Configuration for document chunking behavior.

chunk_mode · enum

Choose how to partition chunks. Options:

• variable - Chunks by character length and visual context
• section - Chunks by section headers
• page - Chunks according to pages
• page_sections - Chunks first by page, then by sections within each page
• disabled - Returns one single chunk
• block - Chunks by individual blocks

chunk_size · integer

The approximate size of chunks (in characters) that the document will be split into. Defaults to null, in which case the chunk size is variable between 250-1500 characters.

ChunkingConfig

Alternate chunking configuration (similar to Chunking).

chunk_mode · enum

Choose how to partition chunks. Options:

• variable - Chunks by character length and visual context
• section - Chunks by section headers
• page - Chunks according to pages
• page_sections - Chunks first by page, then by sections within each page
• disabled - Returns one single chunk
• block - Chunks by individual blocks

chunk_size · integer

The approximate size of chunks (in characters) that the document will be split into. Defaults to None, in which case the chunk size is variable between 250-1500 characters.

EnrichConfig

Configuration for content enrichment using AI models.

enabled · boolean

If enabled, a large language/vision model will be used to postprocess the extracted content. Note: enabling enrich requires tables be outputted in markdown format. Defaults to False.

mode · enum

The mode to use for enrichment. Options:

• standard - Standard enrichment (default)
• page - Page-level enrichment
• table - Table-level enrichment

prompt · string

Add information to the prompt for enrichment.

Spreadsheet

Configuration for spreadsheet processing.

clustering · enum

In a spreadsheet with different tables inside, controls splitting behavior. Options:

• accurate - Applies more powerful models for superior accuracy, at 5× the default per-cell rate
• fast - Default clustering mode
• disabled - Disables clustering; registers as one large table

exclude · List[enum]

Whether to exclude certain elements in the output. Options:

• hidden_sheets
• hidden_rows
• hidden_cols
• styling
• spreadsheet_images

include · List[enum]

Whether to include certain elements in the output. Options:

• cell_colors
• formula

split_large_tables · SplitLargeTables

Configuration for splitting large tables.

SplitLargeTables

enabled · boolean

If True, split large tables into smaller tables. Defaults to True.

size · integer

The size of the tables to split into. Defaults to 50.

ArrayExtractConfig

Configuration for array extraction of long lists.

enabled · boolean

Array extraction allows you to extract long lists of information from lengthy documents. It makes parallel calls on overlapping sections of the document.

mode · enum

The array extraction version to use. Options:

• auto - Automatically selects the best mode
• legacy - Legacy extraction mode
• streaming - Streaming extraction mode
• no_overlap - No overlap between segments

pages_per_segment · integer

Length of each segment, in pages, for parallel calls with array extraction.

streaming_extract_item_density · integer

Number of items to extract in each stream call. Lower numbers will increase quality but be much slower. 50 works well for most documents with tables.

ParseOptions

Complete parsing options combining all configuration types.

enhance · Enhance

Enhancement configuration.

formatting · Formatting

Formatting configuration.

retrieval · Retrieval

Retrieval configuration.

settings · Settings

General settings configuration.

spreadsheet · Spreadsheet

Spreadsheet-specific configuration.

WebhookConfigNew

Configuration for webhook delivery.

channels · List[string]

A list of Svix channels the message will be delivered down. Omit to send to all channels.

metadata · object

JSON metadata included in webhook request body.

mode · enum

The mode to use for webhook delivery. Options:

• disabled - No webhook delivery (default)
• svix - Use Svix for webhook delivery (recommended for production)
• direct - Direct webhook delivery

url · string

The URL to send the webhook to (if using direct webhook).