Outposts (on-premise)
Making documents searchable that must not leave your building — what an Outpost is, what you manage here, and what happens on the machine.
An Outpost is a machine in your network that makes your documents searchable. You install a meinGPT program on it and point it at folders. The program reads the documents and builds a search index — both stay on that machine, and your files are not copied to the cloud.
You only need an Outpost when documents must not leave the building. Every other data source is connected directly under Settings → Data sources, with no hardware of your own.
On-premise does not mean nothing leaves
Every answer takes at least the matching passage out of your network — otherwise there would be no answer. Exactly what leaves is set per document collection via the release level. There are three, and the lowest is still egress.
What you manage where
The most common confusion on first contact: there are two surfaces, split by the question of who makes the decision.
| In meinGPT (Settings → On-premise) | In the program on the Outpost machine |
|---|---|
| which Outposts exist, whether they are connected, when last | which folders and drives are read |
| who may find which documents and internal applications | when they are read, where the search index lives |
| how much an Outpost may release | diagnosing errors, rebuilding the search index |
| revoking a machine's access | — |
In short: here you decide who may see what and what may leave the building. On the machine you decide what gets read at all.
The three release levels
Each level includes everything from the one above it.
| Level | Leaves your network | What it is for |
|---|---|---|
| Passages | the matching paragraphs and the file name | Maximum restraint. Answers feel fragmented, because the model never sees a document in context. |
| Full text (default) | plus the complete text of the documents found | The quality step. The file itself stays with you. |
| Original files | plus the file itself | People can open the file from a citation and reuse it. For a quote template, the file is the answer. |
The difference between the last two is text versus artefact, not machine versus human: once you release the full text, anyone can have the assistant reproduce the content. What stays protected is the form — the layout, the signed PDF, the branded template.
What to know before the rollout
Windows permissions do not carry over
Releasing a folder releases it to everyone who may see that document collection in meinGPT — including people who could not open it in Explorer. NTFS and Active Directory permissions are not inherited. Mapping folders to teams is therefore its own decision, and one you have to make deliberately.
When the machine is off, meinGPT keeps working. Only this Outpost's documents are unfindable for that time — people get answers without those sources, not no answers. An Outpost should still run continuously, otherwise your documents are missing at night and on weekends.
A machine always calls outward. No port is opened in your firewall; the Outpost establishes the connection itself. What it needs is outbound HTTPS to *.meingpt.com.
Revoking access disconnects a machine immediately — without its cooperation. That is the control for "the laptop is gone" or "that employee left". The processed data stays on the machine; it becomes reachable again only after being reconnected.
Allowances
During the beta each workspace has default limits, visible on the on-premise page:
- 1 GB total size of processed source files across all Outposts
- 1 internal application (ERP, internal API, ticket system, …)
Both can be raised — talk to us; the on-premise page shows the way there as soon as a limit is reached.
Both limits are always enforced. The application limit at create time. The storage allowance blocks only above 110% of the number — there is headroom below that. Once exceeded, semantic search on your Outposts is blocked until there is room again or the allowance is raised. If usage cannot be determined because an Outpost is unreachable, nothing is blocked.
Data sources in meinGPT Cloud have their own allowance — 300 GB by default, enforced the same way. It is a guardrail rather than a package: most workspaces hold a fraction of it.
Setup
Windows is the only path you can take yourself during the beta; Docker on your own Linux server we set up together.