Launching today

Papiro
The self-hosted document archive that's 182× lighter
2 followers
The self-hosted document archive that's 182× lighter
2 followers
Papiro is an open-source, self-hosted document archive and a lightweight alternative to Paperless-ngx. Drop in PDFs, photos or Office files and it OCRs them, so any receipt, contract or ID is one search away, in under a millisecond. One container that idles at 5 MB of RAM (Paperless-ngx: 931 MB across 5). Imports a Paperless archive in one command, suggests folders and tags without any AI service, installs on your phone, and has a REST API and an MCP server. MIT licensed.







Free
Launch Team

Render Sandboxes: a computer for every agentRun untrusted code safely. Idle sandboxes cost nothing
Promoted
Hey Product Hunt 👋
I keep all my paperwork (IDs, contracts, receipts, warranties) in a self-hosted Paperless-ngx. It's a great project, but on my small VPS it was using ~480 MB of RAM doing nothing, across five containers, for one person's documents. So I rebuilt the idea from scratch, for people like me.
Papiro is one Rust binary with SQLite inside. Same Tesseract OCR, same "drop a PDF, find it later" workflow, a fraction of the machine. Measured side by side on the same box with the same documents:
182× less RAM at idle (5 MB vs 931 MB with 10,000 documents)
35× smaller install (149 MB vs 5.16 GB of images)
49× faster ingest of digital PDFs, 10× faster with OCR
196× faster lookups among 10,000 documents
Every number has a script and raw data in the repo, including where Paperless still wins (PDF/A, deskewing, workflows, 37 languages). I'd rather you trust the numbers than be impressed by them.
What you get
Search as you type, with filters by folder, tag, correspondent and date
Nested folders, drag and drop, bulk edit
Local auto-filing: it suggests folder, type and tags from documents you already filed, and spots expiry dates and totals. No AI service, nothing leaves your server
A "due dates" view for passports, insurance and leases
An installable phone app with camera capture
Crash-safe ingest, a daily SHA-256 check of every file, backups with 14 days of history
REST API with OpenAPI, signed webhooks, and an MCP server so Claude can search and file for you
papiro import brings a whole Paperless archive over without redoing OCR
Try it
Live demo, no install: getpapiro.pages.dev/demo
Run it: docker run -p 8000:8000 -v papiro-data:/data ghcr.io/elpablazo/papiro (amd64 + arm64)
Full disclosure: I built it with Claude Code as my pair programmer, and every claim above was tested, not just written.
I'd love your feedback: what would make you switch from what you use today? And if you're on Paperless, tell me what you'd miss most. I'll be here all day answering. 🙏