Runs on a private AI box you own
One compact GB10-class device — 128 GB unified memory — runs it on-premise, sold and provisioned by us. No cloud API keys, no data leaving the building.
Run LibreChat on a private AI box
License: MIT
A self-hosted, multi-user chat interface — agents, MCP tools, and document Q&A over a model that lives on your own hardware.
What it is
- An open-source ChatGPT-style web app for teams: accounts, conversations, presets, and file uploads.
- Supports agents, MCP tools, and retrieval over your own documents.
- Runs entirely self-hosted — the UI and its data stay on infrastructure you control.
LibreChat is an independent project. BrainOutput has no affiliation, endorsement, or partnership with it — we sell the private hardware you can run it on. Project: LibreChat.
Cloud API keys vs your private box
The tool is the same either way. What changes is where your prompts and files go, and how you pay.
| Cloud API keys | Your private box | |
|---|---|---|
| Data location | Prompts and files are sent to a third-party API | Everything stays on hardware in your office |
| Cost model | Metered per token — scales with usage, indefinitely | A fixed hardware cost — local inference is unmetered |
| Availability | Vendor uptime, rate limits, and model retirements apply | Runs offline; you control uptime and model versions |
| EU compliance | Data-transfer and processor terms to manage per vendor | EU hardware on your premises — sovereignty is physical, not contractual |
| Model choice | Whatever the vendor exposes | Any open-weight model that fits the box — swap freely |
Indicative hardware start prices are €3 650,22 HT (ASUS) / €4 328,25 HT (Dell); the full sovereign setup is sur devis.
How it connects to your box
LibreChat reaches any OpenAI-compatible server through its custom-endpoints configuration.
- In librechat.yaml, add a custom endpoint with the base URL of the model server running on your box.
- Name the served model and you are done — no external key required for local inference.
How BrainOutput fits
- We provision the GB10-class device and the OpenAI-compatible model stack LibreChat points at.
- You get one private endpoint for the whole team, sized to the models you need.
Frequently asked questions
- Do I still need an OpenAI key?
- No. Once LibreChat points at the local endpoint on your box, inference runs on your hardware with no third-party key or per-token cost.
- Can the whole company use one box?
- Yes — LibreChat is multi-user by design. We size the device (and stack units if needed) for your concurrency.
- Where does our chat data live?
- On infrastructure you control. LibreChat is self-hosted and the model runs on your premises, so conversations never leave the building.
Run it on a box we set up for you
We size the device, install an OpenAI-compatible model stack, and hand you the endpoint this app points at. The app stays free and yours.