> For the complete documentation index, see [/llms.txt](/llms.txt).

# Document Parse

> Source: /docs/models/document-parse

Earlier versions, including deprecated ones, are listed in [All models](/docs/models/history#document-parse).

Make your documents AI-readable

[Compare](/docs/models/compare/document-intelligence) · [Try in Playground](/playground/document-parsing)

- **Max pages per request**: 100 pages (up to 2000 with async API)
- **Supported character sets**: A · 가 · あ · ア · 漢
- **Outputs**: HTML · Markdown
- **Price**: $0.01/page (Standard)
- **Price**: $0.03/page (Enhanced)

The ultimate product for transforming complex documents into formats that Large Language Models (LLMs) can seamlessly process.

- 0.6 page/s on average
- 5–10x faster than competitors
- 2,000 pages via Async API
- [Auto mode for cost & accuracy](/api/parse/document-parsing)

## Supported file formats

- JPEG: Supported
- PNG: Supported
- BMP: Supported
- TIFF: Supported
- HEIC: Supported
- PDF: Supported
- DOCX: Supported
- PPTX: Supported
- XLSX: Supported
- HWP: Supported
- HWPX: Supported

## Maximum file size

100MB (synchronous API) / 500MB (asynchronous API)

## Maximum number of pages per file

Synchronous API: **100 pages** (Files exceeding 100 pages are rejected with a 413 error)

Asynchronous API: **2000 pages**

## Maximum pixels per page

**200,000,000 pixels**

(For non-image files, the pixel count is determined after converting to images at a standard of 150 DPI.)

## Supported character sets (if using OCR)

- Alphanumeric: Input and output
- Hangul: Input and output
- Hiragana: Input and output
- Katakana: Input and output
- Hanja: Input and output
- Hanzi (Beta): Not fully supported
- Kanji (Beta): Not fully supported

Hanja, Hanzi, and Kanji are writing systems based on Chinese characters used in Korean, Chinese, and Japanese writing systems. Despite sharing similarities, they possess distinct visual representations, pronunciations, meanings, and usage conventions within their respective linguistic contexts. For more information, see [this article](https://en.wikipedia.org/wiki/Hanja).

For best results, follow these guidelines:

- Use high-resolution documents to ensure legibility of text. Ensure a minimum document width of 640 pixels.
- The performance of the model might vary depending on the text size. Ensure that the smallest text in the image is at least 2.5% of the image's height.
- For example, if the image is 640 pixels tall, the smallest text should be at least 16 pixels tall.

## Available Platforms

- Upstage Console
- [AWS Marketplace](https://aws.amazon.com/marketplace/pp/prodview-lv5bnpdco7xoq?sr=0-1&ref_=beagle&applicationId=AWSMPContessa)
- [Snowflake Marketplace](https://app.snowflake.com/marketplace/providers/GZTHZ9AB8Y/Upstage)
- [On-premises](https://www.upstage.ai/contact-us?utm_medium=console&utm_source=docs-models&utm_content=text-link)

## Versions

### document-parse-260930 (Alias: document-parse)

Released on 2026-09-30

This release includes the following updates.

- On `document-parse-260930`, `enhanced` mode now uses an end-to-end vision-language model (VLM) parser in place of the previous multi-stage refinement pipeline, and `auto` mode sends complex pages to this parser. The `document-parse` alias now points to `document-parse-260930`, so `enhanced` mode on the alias uses this parser.
- For PDFs with a text layer (digital-born documents), text that the VLM parser recognizes is cross-checked against that text layer to reduce misrecognized characters.
- Richer HTML and Markdown output in `enhanced` mode: heading levels `<h2>`–`<h6>` (depth in `sub_category`), `<ul>`/`<ol>` lists, code blocks with their language, inline text styling, checkboxes, and line breaks within paragraphs.
- Figure and chart captions in `enhanced` mode are written in the language of the document’s intended readers, which the model detects automatically. Korean, English, and Japanese are currently supported.
- Inline equations within text are recognized in `enhanced` mode and returned as LaTeX wrapped in `$...$`.
- Improved recognition accuracy for block equations (the `equation` category) in `enhanced` mode.
- Improved table accuracy and stability on dense pages in `enhanced` mode.
- Available platforms: Upstage Console, [On-premises](https://www.upstage.ai/contact-us?utm_medium=console&utm_source=docs-models&utm_content=text-link)

## API privacy

Data not collected

Upstage does not collect any data from this API

1. API input data is not used for model training.
2. API input data is not stored, unless required for service delivery and explicitly stated.

Free

Try Document Parse in the Playground!

✓ Convert documents to HTML or Markdown

✓ Use a sample or your own file

✓ No API key needed - free to try

[Try it in the Playground](/playground/document-parsing)