ToolsRecipesCompareCategoriesDGX SparkModelsPeopleQuestionsSubmitAboutSponsorNewsletter

← All categories

Local RAG for scanned documents and OCR

Tools tagged for scans and OCR, or whose listing says they read scans/images. 13 tools, sorted by GitHub stars.

📬 Get the weekly list of new local RAG tools and releases. Free, one email a week.

ToolLicenseHardwareDocumentsFully offline
RAGFlow
Active Difficulty: Medium Community report		Apache-2.04 cores, 16GB RAM, 50GB disk recommended (Docker)PDF, scans, DOCX, XLSX, PPT, images✅ Yes
Docling
Active Community report		MITCPU OK, GPU fasterPDF, DOCX, PPTX, XLSX, HTML, EPUB, images, audio, email (EML/MSG), LaTeX✅ Yes
Cherry Studio
Active Difficulty: Easy Docs only		AGPL-3.0Desktop app (Windows, macOS, Linux)PDF, Office docs, images, text (OCR options)✅ Yes
Paperless-ngx
Active Docs only		GPL-3.0Low (Raspberry Pi OK)Scans, PDFs, images, email✅ Yes
LightRAG
Active Docs only		MITDepends on chosen LLM; graph extraction wants a 14B+ modelText, PDF, DOCX; images/tables/Office via RAG-Anything✅ Yes
Khoj
Active Difficulty: Medium Docs only		AGPL-3.0Server; GPU optionalPDF, Markdown, Org, Word, Notion, images✅ Yes
Kotaemon
Active Difficulty: Medium Docs only		Apache-2.0Docker; GPU optionalPDF, HTML, MHTML, XLSX in lite image; DOCX and more need the full image (Unstructured)✅ Yes
MaxKB
Active Difficulty: Medium Docs only		GPL-3.0Docker (PostgreSQL + pgvector)Uploaded documents and crawled web docs; text, image, audio and video input✅ Yes
SurfSense
Active Difficulty: Easy Docs only		Apache-2.0Desktop app (Windows x64, macOS Apple Silicon, Linux x64); pick a local or remote modelPDF (incl. scanned), DOCX, XLSX, PPTX, CSV, images (OCR), Markdown, text, HTML, audio recordings (transcribed locally)✅ Yes
Unstructured (OSS)
Active Docs only		Apache-2.0CPU; hi_res needs more25+ formats incl. scans, email✅ Yes
h2oGPT
Stale (no release in 12+ months) Docs only		Apache-2.0GPU recommendedPDF, images, audio, DOCX, XLSX✅ Yes
R2R
Stale (no release in 12+ months) Difficulty: Medium Docs only		MITDocker, PostgresPDF, DOCX, images, audio⚠️ Needs config
paperless-gpt
Active Docs only		MITNeeds Paperless-ngx + LLMScanned docs✅ Yes

How we know: Docs only = taken from the project's own README/docs, not tested by us. Community report = an owner published a working (or problem) report; links to the source. Vendor guide = an official NVIDIA playbook exists. We have not hands-on tested any tool yet, so nothing here says "Tested by RAGShelf".

Setup recipes

Comparisons

Groupings use only what each listing's data supports. Last updated 2026-10-26.

The RAGShelf weekly

New local RAG tools, releases, and tested setups. One email a week, no spam.