# SoMarkAI/OXR

> OXR is an open-source document parsing model that turns PDFs and document images into Markdown or JSON with layout coordinates, recognizing 13 element types such as tables, formulas, chemical structures and code. With 1.3B parameters it ranks first on olmOCR-Bench, OmniDocBench and DocParsingBench.

- Magnitude: 7.5 out of 10 — Breakout
- Stars: 1,934 total · +52 stars measured 2026-10-28, 03:32–05:40 UTC, ≈ 492 by evening
- Star trust: star growth looks organic
- Category: Language models · Language: Python · Created: 2026-10-05 · Last push: 2026-10-24
- GitHub: https://github.com/SoMarkAI/OXR · Homepage: https://somark.ai/oxr · Page: https://gitnova.dev/en/r/SoMarkAI/OXR

## Useful for

- Parse a PDF or scan into Markdown/JSON via the free API without a key
- Swap the built-in layout model for a custom one on non-standard documents
- Extract tables, formulas and chemical structures with coordinates for an agent pipeline

## Why it’s here

- Star-counter measurements on 2026-10-28 (UTC), 03:32–05:40: 1882 → 1934 stars (+52). This is the change over that interval.
- Estimated end-of-day forecast: about +492 stars, using observed gains and the previous day.
- Over the last two days the pace is 8.4× that of the previous week and a half.
- Elevated gains were measured on the last 3 completed UTC days in a row. Continued growth is not guaranteed.
- The repository is 23 days old and already has 1,934 stars.
- Top new repositories this month: #97.
- About 97 forks a day — people are taking the code.

## Star trust

Star growth looks organic.

Star-trust labels are heuristics based on the repository’s behavior, not a check of every stargazer.

## Numbers

- Forks: 123
- Issues and pull requests: 17
- Watchers: 47
- Average over the last week: 329 per day
- Usual pace: 2 per day
- Stars in the last hour (measured): 17

## Stars per day, last 24 days (oldest → newest, today is partial)

2026-10-05 … 2026-10-28: 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 24, 102, 129, 131, 218, 181, 452, 702, 52

## Spotted in now

- Top new repositories this month: #97

## Similar by description

1. **datalab-to/indra** — 3.4 · Steady · Data & analytics · Python · +27 stars measured 2026-10-28, 00:04–05:42 UTC, ≈ 96 by evening
   A 1M-parameter CPU OCR model that turns PDFs and images into markdown, plain text, or JSON with boxes, layout labels, and reading order. Runs on ONNX Runtime without PyTorch, up to 49 pages/s.
   Full card: https://gitnova.dev/en/r/datalab-to/indra.md

---

Magnitude (0–10) measures how fast and how unusually interest in a repository is growing right now. It is not a quality score.
Days are UTC. “So far today” is a fact; “expected by the end of the day” is a forecast.
Summaries and use cases are written by an LLM (DeepSeek V4.1 Flash) from the README and may be inaccurate: verify specific claims (benchmarks, speed, hardware) in the repository itself.
Data as of 2026-10-28 05:51 UTC, updated every 30 minutes.