# Chunkr

- Category: AI platforms
- Official site: https://lumina.sh/
- HTML: https://www.88z.ai/en/tools/chunkr
- Markdown: https://www.88z.ai/en/tools/chunkr.md

Transform complex documents into high-quality, LLM-ready structured data

## What is Chunkr?

Chunkr is a Document Intelligence API designed for parsing, data extraction, and document pipelines.It efficiently converts complex file formats such as PDFs, images, and spreadsheets into HTML, Markdown, or JSON formats that are ready for Large Language Models (LLMs).

By integrating OCR, layout detection, reading order recognition, bounding box localization, and citation extraction, Chunkr ensures that data extracted from unstructured documents is both accurate and contextually relevant.Additionally, it supports schema-based data extraction, helping developers build more robust document processing pipelines.

## Key features

- **Multi-format Par**：Parses PDFs, images, and spreadsheets into LLM-friendly HTML, Markdown, or JSON formats.
- **Advanced OCR**：Built-in optical character recognition accurately extracts text content, even under complex backgrounds or layouts.
- **Layout & Order D**：Automatically detects document layout and determines the correct reading order to maintain logical coherence.
- **Bounding Box Loc**：Provides precise bounding box coordinates, making it easy to highlight or locate specific data snippets in the original document.
- **Citation Extract**：Automatically identifies and extracts citations from documents, enhancing data traceability and credibility.
- **Schema-based Ext**：Allows users to define data structure schemas to precisely extract fields and data that meet specific requirements from documents.

## How to use it

1. 1.Sign up for a Chunkr account and obtain your API key.
2. 2.Prepare the document files you need to process (PDFs, images, or spreadsheets).
3. 3.Call the Chunkr API, upload the document, and specify the desired output format (HTML/Markdown/JSON).
4. 4.(Optional) Configure schema-based extraction rules to retrieve specific data fields.
5. 5.Receive and process the returned structured data, integrating it into your application or LLM workflow.

## Where it fits

- **RAG Knowledge Ba**：Clean and convert large volumes of unstructured documents into structured data to enhance knowledge retrieval for Generative AI.
- **Financial Docume**：Automatically extract key amounts, dates, and entity information from invoices, receipts, and financial statements.
- **Legal Document A**：Parse contracts and legal files to extract clauses, citations, and key obligations, assisting in legal review processes.

## Pricing and free quota

Pricing and any free quota follow the official site.

## Who it is for

- AI Application Developers
- Data Engineers
- Enterprise Automation Teams
- Legal and Financial Analysts

## FAQ

### Which file formats does Chunkr support?

Chunkr supports parsing PDFs, image files, and spreadsheets.

### What output formats are available?

You can choose to convert documents into HTML, Markdown, or JSON formats to better fit LLM inputs.

### What is schema-based extraction?

Schema-based extraction allows you to predefine data structures, and Chunkr will precisely capture corresponding field values from documents, which is suitable for standardized data entry scenarios.

## Tags

Document Parsing · OCR · Data Extraction

Features, pricing and availability follow the official site. Except for 88z.ai’s own API pages, this is a third-party product; 88z.ai only provides a directory introduction.