# icereed/paperless-gpt

Use LLMs and LLM Vision (OCR) to handle paperless-ngx - Document Digitalization powered by AI

Repository: https://github.com/icereed/paperless-gpt
Canonical: https://ross.abutalabs.com/products/paperless-gpt
Language: Go
License: MIT
License Family: permissive
Topics: ai, chatgpt, llm, ollama, paperless, paperless-ngx, mistral, ocr
Last push: 2026-08-26T15:55:04+00:00

## Health v2 (maintenance only)
Score: 87/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 99, release rhythm 94, longevity 50
- inputs: {"age_days": 709, "days_push": 7, "days_rel": 43, "gap_med": 5.5, "n_releases_24m": 47}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 2651, forks 202 (observed 2026-08-28T04:07:07.024121+00:00)

## What it is
A self-hosted companion application for paperless-ngx that uses LLMs and vision models to auto-generate document titles, tags, and dates, and to deliver higher-accuracy OCR on scanned documents. It supports multiple backends including OpenAI, Ollama (including reasoning models), Google Document AI, Azure Document Intelligence, and Docling Server.

## Use cases
- auto tag and title documents in paperless-ngx with AI
- OCR messy low-quality scans with an LLM
- automatically rename and categorize digitized documents
- run local document OCR with Ollama for privacy
- generate document metadata like created dates with ChatGPT or Mistral
- batch process scanned PDFs into searchable text
- review and refine AI-suggested document classifications

## When to choose
- You already run paperless-ngx and want AI-generated titles, tags, and dates for your documents
- You need better-than-traditional OCR on tricky or low-quality scans
- You want to keep document processing local and private using Ollama or a self-hosted Docling server
- You want enterprise OCR options like Google Document AI or Azure Document Intelligence wired into paperless-ngx

## When to avoid
- You do not use paperless-ngx and need a standalone OCR or document management tool
- You need a full document management system - paperless-gpt is a companion, not a DMS
- You have no access to any LLM API or cannot run a local model
- You need fully hands-off automation with no human review of AI suggestions

## Facets
- artifact type: application
- maturity: active
- function: ocr, machine-learning, llm-inference
- domain: artificial-intelligence, developer-tools, self-hosted, pdf
- platform: self-hosted, cross-platform
- tags: paperless-ngx, document-management, document-digitization, ollama, openai, vision-llm, auto-tagging, title-generation, google-document-ai, azure-document-intelligence, docling, automation, docker, linux

## Member repositories
- icereed/paperless-gpt (main) score 87

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:07:07.024121+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 2026-08-30T02:18:52.334406+00:00, confidence not recorded.
  - readme: https://github.com/icereed/paperless-gpt (fetched 2026-08-28T04:07:07.024121+00:00, sha e3922f403ca7)
- Data as of 2026-08-30T08:39:29.467469+00:00.