SoMarkAI/OXR
An open-source document parsing model with SOTA accuracy and speed, using an RSI loop during post-training.
From the publisher
README & documentation
Source previewRead the project’s overview, installation instructions and usage examples. The original README is the source of truth.
SoMark｜OXR An open-source document parsing model with SOTA accuracy and speed, using an RSI loop during post-training. Turn documents into Agent-Friendly data. 🌐 Online Demo ｜ 🔥 🆓 Free API [](https://github.com/SoMarkAI/OXR) [](https://huggingface.co/SoMarkAI/OXR-1.0) [](https://modelscope.cn/models/SoMark/OXR-1.0) English · 简体中文 OXR (Optical Everything Recognition) is an open-source model from the SoMark team. It turns PDFs and document images into Agent-Friendly data in Markdown or JSON, with layout coordinates included. Pretrained from scratch, OXR uses a recursive self-improvement (RSI) loop during post-training—a first in document parsing. With just 1.3B parameters, OXR ranks #1 on olmOCR-Bench, OmniDocBench, and DocParsingBench while delivering industry-leading parsing speed. OXR is the first to make its layout analysis module independent and pluggable. This design accelerates parsing and supports replacing the built-in layout model with a custom one. OXR continues to handle content recognition and reading order, enabling adaptation across diverse scenarios with minimal integration effort. Expand to see all 13 element…
Read the full README ↗ · Preview checked 2026-10-28T12:24:54.680Z
Inside the original README — Document outline
- Why OXR?
- Parsing Examples
- How it works
- Technical Architecture
- Improvement Through the RSI Loop
- Quick Start
- 0. 🔥 🆓 Free API
- 1. Run Locally
- 1.1 Download the Code and Models
- 1.2 Start the Services
- 1.2.1 Run with Docker (Recommended)
- 1.2.2 Run from Source
Headings are captured from the source. Links open the publisher’s document, not a locally hosted copy.
Links from the README
References supplied by the publisher, not independently verified endorsements. Check the destination before downloading files or entering credentials.
Documentation belongs to its respective authors. Reported project/model license: Other. A listing is not a grant of reuse or training rights. Confirm the document’s own terms at the source.
What this repository does
An open-source document parsing model with SOTA accuracy and speed, using an RSI loop during post-training.
Repository facts
- Owner
- SoMarkAI
- Primary language
- Python
- Stars
- 2,721
- Forks
- 195
- Open issues + pull requests
- 21
- License
- Other
- Archived
- No
- Default branch
- main
- Created
- 2026-10-05T04:28:31.000Z
- Last push
- 2026-10-24T15:48:22.000Z
Topics and intended use
Owner-supplied topics: agent, llm-tools, ocr, oxr, rsi, vlm, vlm-ocr
Review the README for scope, installation, examples and limitations. We do not run repository code or certify it.
Evaluate before installing
Review licensing and dependencies, inspect recent commits and unresolved issues, and test in an isolated environment before production use. Stars and forks alone cannot answer these questions.
Source and freshness
Source: GitHub. Metadata observed 2026-10-31T12:18:20.650Z. Daily imports are snapshots, not real-time monitoring.
Popularity and source listings do not establish security, suitability, licensing rights or benchmark performance.
Related repositories
EbookFoundation/free-programming-books
donnemartin/system-design-primer