# Ucas-HaoranWei/GOT-OCR2.0

Official code implementation of General OCR Theory:  Towards OCR-2.0 via a Unified End-to-end Model

Repository: https://github.com/Ucas-HaoranWei/GOT-OCR2.0
Canonical: https://ross.abutalabs.com/products/got-ocr20
Language: Python
License Family: other
Last push: 2025-02-10T12:13:00+00:00

## Health v2 (maintenance only)
Score: 25/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 6, release rhythm 35, longevity 52
- inputs: {"age_days": 730, "days_push": 569, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases, no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 8216, forks 711 (observed 2026-08-28T04:10:19.615705+00:00)

## What it is
Official implementation of GOT-OCR2.0, a unified end-to-end vision-language model for general OCR that converts images of text, documents, formulas, and charts into structured text. It is a research codebase with a Hugging Face model, now also integrated into the transformers library.

## Use cases
- extract text from images with ocr
- convert scanned pdf documents to markdown
- recognize math formulas from screenshots
- parse tables and charts into text
- run ocr with a vision language model
- fine-tune an ocr model on my own data

## When to choose
- you need end-to-end OCR beyond plain text, including formulas, tables, and charts
- you want a Hugging Face transformers-compatible OCR model
- you are doing document-parsing research or building document ingestion pipelines

## When to avoid
- you need a lightweight CPU-only OCR like Tesseract
- you require a permissive license - the repo has no license, limiting commercial use
- you need production support or a stable API rather than research code

## Facets
- artifact type: library
- maturity: active
- function: ocr, machine-learning, deep-learning, llm-inference, image-processing, pdf
- domain: computer-vision, artificial-intelligence, pdf, deep-learning
- platform: python, cross-platform
- tags: ocr-2-0, vision-language-model, end-to-end-ocr, document-parsing, research-code, no-license, natural-language-processing, gpu, linux

## Member repositories
- Ucas-HaoranWei/GOT-OCR2.0 (main) score 25

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:10:19.615705+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T17:29:31.996001+00:00, confidence not recorded.
  - readme: https://github.com/Ucas-HaoranWei/GOT-OCR2.0 (fetched 2026-08-28T04:10:19.615705+00:00, sha 4fccb476f76f)
- Data as of 2026-08-30T08:39:29.467469+00:00.