Open-source 1.3B model OXR-1.0 tops 3 doc-parsing benchmarks and is the fastest, using an RSI loop with no human labels

rohanpaul_ai · x · 2026-10-28

SoMark open-sourced OXR-1.0, a 1.3B-parameter document parsing model that turns PDFs and page images into Markdown or JSON.

Results

Architecture

Recursive self-improvement (RSI) loop

Related event: SoMark Open-Sources OXR-1.0, a 1.3B Document Parsing Model Hitting SOTA(2 posts)→

Original post →

More from Models

Models channel →