#3939th of 50 in Data and scraping for AIFor data and documents, documents and ocr
oxr
1.3B document parsing model that turns PDFs and images into Markdown or JSON
- Stars
- 3.8k
- License
- custom license
- Last commit
- Nov 2026
- Language
- Python
10 on 6 Oct 20263,772 on 10 Nov 2026
From GitHub's daily star counts.
Checked on GitHub.
Overview
OXR is a 1.3B-parameter model from SoMark that parses PDFs and document images into Markdown or JSON with layout coordinates, covering 13 element types such as formulas, tables, chemical structures and code blocks. Layout analysis runs as a separate, replaceable model, while OXR handles recognition and reading order. It serves a SoMark-compatible parsing API and a demo page on port 8000, using vLLM for the model.
Who it is for: Engineers self-hosting document parsing for RAG or agents on NVIDIA GPUs
Strengths
- Layout model is swappable while OXR keeps recognition and reading order
- Outputs Markdown or JSON with bounding boxes for 13 element types
- API works with somark-py and somark-js CLIs by changing the base URL
- Local demo at /demo plus a free hosted API for trying it
Weaknesses
- Needs Linux and an NVIDIA BF16 GPU; 24GB VRAM minimum, 48GB recommended
- Custom license based on Apache 2.0 with extra terms
- Official vLLM lacks OXR support; an out-of-tree plugin is required
- Benchmark rankings are self-reported; technical report not yet published
What it needs
- GPU required
- Docker
- Needs NVIDIA driver (CUDA 13.0 compatible), vLLM 0.25.1, PyTorch 2.11.0, Python 3.12, Docker
- Models: OXR-1.0 (BF16, 1.3B), D-FINE layout analysis model
- port 8000
Quick answers
Is oxr open source?
Its code is public on GitHub under its own license terms; read them before you use it at work.
Can you run oxr with Docker?
Yes. Its repository has a Docker setup: a Dockerfile, a Compose file or a published image.
Does oxr need a GPU?
Yes, it needs a GPU.
Also in Data and scraping for AI
See all 50| Rank | Project | Score |
|---|---|---|
| 1 | doclingPython library that converts PDFs, Office files and more into structured text | 92 out of 100 |
| 2 | mineruParses PDFs, images and Office files into Markdown and structured JSON | 91 out of 100 |
| 3 | opendataloader-pdfPDF parser that outputs Markdown, JSON with bounding boxes, and Tagged PDF | 90 out of 100 |
| #4 | agent-reachCLI that gives coding agents read access to login-gated and Chinese web platforms | 88 out of 100 |
| #5 | markitdownPython library and CLI that converts office files, PDFs and audio to Markdown | 87 out of 100 |