Overview
Document Parse AI Solution for Unstructured Document Parsing DocumentAI is an AI powered document parsing solution that transforms various unstructured documents into structured data. It supports a wide range of formats including PDF, HWP, PPTX, DOCX, and more. It precisely analyzes not only text but also tables, formulas, code, images, and other elements within documents. The parsed output is provided in a standard JSON format, arranged according to the natural reading flow, making it easy for developers to use in AI training, search systems, automation tasks, and more. Reduce preprocessing time and maximize the value of unstructured data with DocumentAI.
Highlights
- Key Strengths Accurate Layout Analysis: Automatically detects document structure and provides high-quality structured data. Ensures consistent parsing across diverse formats and layouts. Wide Format Support: Supports PDF, DOCX, HWP, PPTX, XLSX, and image files. Enables unified processing of documents from various sources. JSON-Based Output: Delivers human-readable JSON results. Easily integrated into AI training, search, and automation systems.
- Key Features Text OCR: Extracts text from images or PDFs, including metadata-based parsing. Layout Analysis: Classifies content into 10 types (text, tables, formulas, code, etc.). Table Parsing: Converts tables, including image-based ones, into structured HTML. Formula Parsing: Extracts formulas from tables or standalone and converts to LaTeX. Text Parsing: Segments and organizes text by semantic units. JSON Output: Provides all results in a unified, readable JSON format.
- Key Applications LLM Preprocessing: Structures large volumes of documents to build quality training datasets. RAG System Development: Parses various formats to construct knowledge bases for RAG systems. Enterprise KMS: Converts unstructured documents (e.g., reports, manuals) into searchable, reusable data. Automated Report Generation: Generates formatted reports automatically using parsed data.
Details
Unlock automation with AI agent solutions

Features and programs
Financing for AWS Marketplace purchases
Pricing
Dimension | Cost/hour |
---|---|
g6.xlarge Recommended | $15.00 |
Vendor refund policy
We do not support any refunds currently.
How can we make this page better?
Legal
Vendor terms and conditions
Content disclaimer
Delivery details
64-bit (x86) Amazon Machine Image (AMI)
Amazon Machine Image (AMI)
An AMI is a virtual image that provides the information required to launch an instance. Amazon EC2 (Elastic Compute Cloud) instances are virtual servers on which you can run your applications and workloads, offering varying combinations of CPU, memory, storage, and networking resources. You can launch as many instances from as many different AMIs as you need.
Version release notes
Added enhanced security features to improve overall product protection.
Additional details
Usage instructions
Input Supported File Formats: PDF, DOCX, PPTX, HWPX, HWP
Upload Methods: Files can be uploaded individually through the web interface or automatically via API.
Input data
POST /api/v1/document/parse
Content-Type: multipart/form-data
Authorization: Bearer YOUR_API_KEY
Request Body:
{
"file": binary,
"title": "File Name",
"subtitle": "Sub Title"
}
Output Format (Structured JSON) The output is provided in a structured JSON format, sorted in a natural reading order.
Top-Level Fields msg_id: Unique ID for server communication
subtitle: Subtitle assigned by the user to the document
title: Actual title of the document
save_time: Timestamp of the document request
result Object api Version of the API used for processing
content html: Entire content of the analyzed document in HTML format
markdown: Entire content in Markdown format
text: Entire content in plain text format
elements (Array of document components) Each element includes:
category: Recognized class name (e.g., plain_text, figure, caption, etc.)
content:
text: The raw textual content of the element
markdown: Markdown representation
html: HTML representation
coordinates:
Four (x, y) coordinates indicating the position of the element
Represent the top-left, top-right, bottom-right, and bottom-left corners
conf: Confidence score of the recognition (float between 0 and 1)
id: Unique ID assigned to the element (integer)
page: Page number where the element appears
order: Reading order of the element on the page (integer)
relation (optional): List of related element IDs (e.g., caption linked to a figure or table)
model Version of the AI model used (e.g., "interx_doc_ai_250509")
pages Total number of pages in the document
Latency_per_module table: Time taken for table recognition (in seconds)
formula: Time taken for formula recognition (in seconds)
text: Time taken for text analysis (in seconds)
end2end_execution_time: Total end-to-end processing time (in seconds)
Output data { "msg_id": string, "subtitle": string, "title": string, "save_time": string, "result": { "api": string, "content": { "html": string, "markdown": string, "text": string }, "elements": [ { "category": string, "content": { "text": string, "markdown": string, "html": string }, "coordinates": [ { "x": float, "y": float }, { "x": float, "y": float }, { "x": float, "y": float }, { "x": float, "y": float } ], "conf": float, "id": integer, "page": integer, "order": integer, "relation": [integer] (optional) } ], "model": string, "pages": integer, "Latency_per_module": { "table": float, "formula": float, "text": float, "end2end_execution_time": float } } }
Support
Vendor support
For inquiries regarding the documentAI model, please contact us through
AWS infrastructure support
AWS Support is a one-on-one, fast-response support channel that is staffed 24x7x365 with experienced and technical support engineers. The service helps customers of all sizes and technical abilities to successfully utilize the products and features provided by Amazon Web Services.
Similar products


