Document Parsing

Parse documents to extract structured content with parse.run()

The Parse API converts documents into structured content, extracting text, tables, images, and layouts with high accuracy.

Basic Usage

Sync

from reducto import Reducto

client = Reducto()

response = client.parse.run(
    input="https://example.com/document.pdf"
)
print(response)

Async

import asyncio
from reducto import AsyncReducto

client = AsyncReducto()

async def main():
    response = await client.parse.run(
        input="https://example.com/document.pdf"
    )
    print(response)

asyncio.run(main())

Method Signature

client.parse.run(
    input: str,
    enhance: Enhance | None = None,
    formatting: Formatting | None = None,
    retrieval: Retrieval | None = None,
    settings: Settings | None = None,
    spreadsheet: Spreadsheet | None = None,
    async_: ConfigV3AsyncConfig | None = None
) -> ParseRunResponse

Parameters

input · string · required

The URL of the document to parse. You can provide:

• A publicly available URL
• A presigned S3 URL
• A reducto:// prefixed URL from the /upload endpoint
• A jobid:// prefixed URL from a previous parse invocation
• A list of URLs (for multi-document pipelines, V3 API only)

enhance · Enhance

Enhancement options for improving extraction accuracy.

Show properties

agentic · array

Uses vision language models to enhance accuracy for tables, figures, or text. Increases cost and latency.

summarize_figures · boolean · default: "true"

If true, summarize figures using a small vision language model.

formatting · Formatting

Control output formatting and structure.

Show properties

add_page_markers · boolean · default: "false"

Add page markers to the output. Useful for extracting data with page-specific information.

table_output_format · enum · default: "dynamic"

Format for table output: html, json, md, jsonbbox, dynamic, or csv. Dynamic returns markdown for simple tables and HTML for complex tables.

merge_tables · boolean · default: "false"

Merge consecutive tables with the same number of columns.

include · array

List of formatting elements to include: change_tracking, highlight, comments, hyperlinks, signatures.

retrieval · Retrieval

Configure chunking for retrieval-optimized output.

settings · Settings

Processing settings and preferences.

spreadsheet · Spreadsheet

Spreadsheet-specific parsing options.

async_ · ConfigV3AsyncConfig

Configuration for asynchronous processing. When provided, the request returns immediately with a job ID.

Advanced Example

Sync

from reducto import Reducto

client = Reducto()

response = client.parse.run(
    input="https://example.com/document.pdf",
    enhance={
        "summarize_figures": True,
        "agentic": ["table", "figure"]
    },
    formatting={
        "add_page_markers": True,
        "table_output_format": "json",
        "merge_tables": False
    }
)

# Access the parsed content
print(response.content)

Async

import asyncio
from reducto import AsyncReducto

client = AsyncReducto()

async def main():
    response = await client.parse.run(
        input="https://example.com/document.pdf",
        enhance={
            "summarize_figures": True,
            "agentic": ["table", "figure"]
        },
        formatting={
            "add_page_markers": True,
            "table_output_format": "json",
            "merge_tables": False
        }
    )

    # Access the parsed content
    print(response.content)

asyncio.run(main())

Async Job Processing

For long-running documents, use run_job() to process asynchronously:

from reducto import Reducto

client = Reducto()

# Start an async job
job = client.parse.run_job(
    input="https://example.com/large-document.pdf",
    async_={"webhook": {"url": "https://example.com/webhook"}}
)

print(f"Job ID: {job.job_id}")

# Poll for results
result = client.job.get(job.job_id)

Input Formats

The Parse API supports multiple input methods:

Direct URL

response = client.parse.run(
    input="https://example.com/document.pdf"
)

File Upload

from pathlib import Path

# First upload the file
upload_response = client.upload(
    file=Path("/path/to/document.pdf")
)

# Then parse using the reducto:// URL
response = client.parse.run(
    input=upload_response.url
)

Reuse Previous Parse

# Use output from a previous parse job
response = client.parse.run(
    input=f"jobid://{previous_job_id}"
)