(2)
Gathers everything, wherever it landed
Your inbox, your drive, the bank export in Downloads. Archive reads them all and keeps each original exactly as it came.
See what it connects toWorks with everything you use(1),
and keeps a copy of all of it.
(1)Connect your inbox, your drives, your bank exports and the apps you already pay for. Archive pulls everything in and keeps every original, untouched.
(3)
Tidies it into records you can trust
A crumpled receipt becomes a clean record. Every value remembers where it came from, and how sure Archive is about it.
See a record
(4)
Ask it anything, from anywhere
Search in the app, call the API, or ask Claude over MCP. Answers come from your own records, with the sources to prove it.
Read about the API
(5)
Private by design
Run it on your own laptop with your own models, or let us host an isolated instance. Either way, your data never trains anyone's AI.
How self-hosting worksWhy Archive
Your receipts, mail and files become one archive that is searchable, portable and yours alone.
(6)
One command. That is the install.
It checks for Docker, starts Archive with everything inside, and keeps your data in one volume. Run it again to update.
curl -fsSL https://archive.us.org/install.sh | sh- Getting ghcr.io/mulholland-inc/archive:latest
- Starting Archive
- Archive is running at http://localhost:8080
- Run this command again any time to update.
One folder holds it all
Your database, files and settings live in /data, so a backup is one folder.
Upgrades are a pull
Fetch the new image and restart. Archive takes a backup before it changes anything.
Bring your own AI
Point Archive at Ollama, llama.cpp, vLLM or LM Studio. Models never ship inside the image.
Already run Postgres?
Flip one setting and Archive uses yours instead of the bundled one.
Run it anywhere.
All it needs is a container runtime, a volume and one port.
| Where | How | Database |
|---|---|---|
| Your laptop or a VPS | Docker or Podman | Bundled |
| Kubernetes | A Helm chart | Bundled, or your own |
| AWS | ECS Fargate or App Runner | RDS or Aurora |
| Google Cloud | Cloud Run | Cloud SQL |
| Azure | Container Apps | Azure Database for PostgreSQL |
| Fly.io, Railway, Render | Native container deploy | Their managed Postgres |
(7)
Free to keep, pay for comfort.
Every plan is the same Archive, so you can move between them and take everything with you. Prices land at launch.
Billed monthly.
Billed yearly, two months free.
Free
Run it yourself on a laptop, a home server or your own cloud.
€0 forever
Self-hosted, nothing held back
- Everything Archive does
- Local AI models
- App, API, MCP and CLI
Personal Easiest start
We run your archive for you. Sign up, connect your stuff, done.
€— /mo
Everything in Free, plus:
- Your own isolated instance
- Updates and backups handled
- Move to self-hosting any time
Team
For households and small teams that share some things and not others.
€— /mo
Everything in Personal, plus:
- Shared and private spaces
- Owner, admin, editor and viewer roles
- Private things stay private
We install it for you
For businesses. We set Archive up on your own systems.
Let us talk
Everything in Team, plus:
- On your cloud or servers
- Single sign-on and provisioning
- Setup kept as code with Terraform
Questions, answered
Is the free version really everything?
Yes. Self-hosted Archive is the whole product. The paid plans are for when you would rather we ran it.
Can I move between plans?
Any time. Your archive is one folder, so moving from hosted to self-hosted, or back, is an export and an import.
Who can see my data on a hosted plan?
Each archive runs in its own isolated instance. We do not read it, and it never trains anyone's AI.
Be first in line.
We will write once, when your archive is ready. Nothing else.