# imanoop7/Ollama-OCR

Repository: https://github.com/imanoop7/Ollama-OCR
Canonical: https://ross.abutalabs.com/products/ollama-ocr
Language: Jupyter Notebook
License: MIT
License Family: permissive
Last push: 2025-03-17T03:10:56+00:00

## Health v2 (maintenance only)
Score: 26/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 11, release rhythm 35, longevity 45
- inputs: {"age_days": 643, "days_push": 534, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 2780, forks 322 (observed 2026-08-28T04:07:21.602757+00:00)

## What it is
A Python package and Streamlit web app that performs OCR on images and PDFs using vision language models served through Ollama. It supports multiple vision models (LLaVA, Llama 3.2 Vision, Granite3.2-vision, Moondream, MiniCPM-V) and output formats like Markdown, JSON, and tables.

## Use cases
- extract text from images using local vision models
- convert scanned pdfs to markdown
- ocr documents without cloud services
- extract tables from images
- batch process images for text extraction
- run ocr with llama3.2-vision or moondream

## When to choose
- you want privacy-preserving, fully local OCR via Ollama
- you need structured output formats like markdown, JSON, or key-value pairs from documents
- you want to experiment with multiple open vision-language models for text extraction

## When to avoid
- you need maximum OCR accuracy on complex scans and prefer mature engines like Tesseract or cloud OCR APIs
- you cannot run local LLMs due to limited hardware
- you need a lightweight dependency-free OCR solution

## Facets
- artifact type: library
- maturity: active
- function: ocr, image-processing, pdf, llm-inference, nlp
- domain: computer-vision, image-processing, pdf, artificial-intelligence, developer-tools
- platform: python, cross-platform, self-hosted
- tags: ollama, vision-language-models, streamlit, text-extraction, batch-processing

## Member repositories
- imanoop7/Ollama-OCR (main) score 26

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:07:21.602757+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 2026-08-30T08:16:07.560335+00:00, confidence not recorded.
  - readme: https://github.com/imanoop7/Ollama-OCR (fetched 2026-08-28T04:07:21.602757+00:00, sha 0336e8d3f0c4)
  - registry_pypi: https://pypi.org/pypi/ollama-ocr/json (fetched 2026-08-29T09:55:32.095201+00:00, sha 7d15e3c3d986)
- Data as of 2026-08-30T08:39:29.467469+00:00.