#1313th of 82 in Data and scraping for AIFor data and documents, documents and ocr

surya

Document OCR, layout and table recognition VLM with a lightweight Python client

Releases and problems feed (RSS)

Stars
21.6k
License
Apache-2.0
Last commit
Nov 2026
Last release
Nov 2026
Language
Python
Stars, last 29 weeks+2,091
28 Apr 2026: 19,515 stars5 May 2026: 19,545 stars12 May 2026: 19,575 stars19 May 2026: 19,608 stars26 May 2026: 19,635 stars2 Jun 2026: 20,382 stars9 Jun 2026: 20,615 stars16 Jun 2026: 20,699 stars23 Jun 2026: 20,770 stars30 Jun 2026: 20,882 stars7 Jul 2026: 20,931 stars14 Jul 2026: 20,979 stars21 Jul 2026: 21,021 stars28 Jul 2026: 21,068 stars4 Aug 2026: 21,101 stars11 Aug 2026: 21,147 stars18 Aug 2026: 21,196 stars25 Aug 2026: 21,226 stars1 Sept 2026: 21,253 stars8 Sept 2026: 21,283 stars15 Sept 2026: 21,313 stars22 Sept 2026: 21,340 stars29 Sept 2026: 21,365 stars6 Oct 2026: 21,408 stars13 Oct 2026: 21,446 stars20 Oct 2026: 21,477 stars27 Oct 2026: 21,512 stars3 Nov 2026: 21,551 stars10 Nov 2026: 21,577 stars17 Nov 2026: 21,606 stars

19,515 on 28 Apr 202621,606 on 17 Nov 2026

From GitHub's daily star counts.

Checked on GitHub.

Datalab is hiring: 4 open roles

Overview

Surya is a 689M-parameter vision-language model that does page OCR, layout analysis with reading order, and table recognition (HTML with colspan/rowspan). The surya-ocr package is a thin client that spawns or attaches to a vllm, llama.cpp or MLX server. It ships CLI commands, a Python API and a local web viewer (surya_gui). Code is Apache-2.0; model weights use a modified OpenRAIL-M license.

Who it is for: Engineers needing self-hosted OCR, layout and table extraction

Strengths

  • One VLM handles OCR, layout, reading order, tables and inline math (KaTeX LaTeX)
  • Backends: vllm on NVIDIA, llama.cpp on CPU, mlx_vlm on Apple Silicon
  • Reports 85.9 on OmniParseBench and 4.5 pages/s on one H100 at 192 DPI
  • Can attach to any OpenAI-compatible server via SURYA_INFERENCE_URL

Weaknesses

  • Model weights are not Apache: free only for research, personal use, and startups under $5M
  • NVIDIA GPU path needs Docker or Podman plus the NVIDIA Container Toolkit
  • Outputs blocks and HTML only; full Markdown or RAG conversion needs the separate Marker project
  • Self-hosted servers must return special tokens (llama-server needs --special)

What it needs

  • GPU optional
  • Needs vllm (Docker/Podman + NVIDIA Container Toolkit), llama.cpp llama-server, mlx_vlm
  • Models: Surya 2.5 (689M VLM), OpenAI-compatible servers
  • port 8000

What changed

From the release notes. No flag means the note does not say so, not that the upgrade is safe.

Known problems

Open now, reported in the last 90 days

Titles are the reporters' own words, taken from the project's GitHub issues and ranked by reactions. We have not tried to reproduce any of them.

Also in Data and scraping for AI

See all 82
Also in Data and scraping for AI
RankProjectScore
1opendataloader-pdfPDF parser that outputs Markdown, JSON with bounding boxes, and Tagged PDF30.3k stars+662 (+2.2%) in 30 daysApache-2.092 out of 100
2mineruParses PDFs, images and Office files into Markdown and structured JSON Live demo ↗ (opens in a new tab)84k stars+2,037 (+2.5%) in 30 dayscustom license91 out of 100
3doclingPython library that converts PDFs, Office files and more into structured text70.8k stars+1,686 (+2.4%) in 30 daysMIT91 out of 100
#4agent-reachCLI that gives coding agents read access to login-gated and Chinese web platforms113.3k starsMIT88 out of 100
#5markitdownPython library and CLI that converts office files, PDFs and audio to Markdown197k stars+5,594 (+2.9%) in 30 daysMIT87 out of 100