3 bundles. One local AI stack.
You can get the rzfz.ai Stack — the integrated open-source stack for local AI infrastructure — in two ways: as a bundle with matching hardware or a cloud instance, set up turnkey, or as a pure software install on your own hardware. The bundles below show the most common combinations.
One stack. Many shapes.
Space pauses, arrow keys jump between scenes — once the player has focus.
Film transcript
- Cold open
One stack. Many shapes. rzfz.ai: local AI on the hardware you choose
- One box: redaction
Everything on one box. Docling reads, Presidio and the local LLM find personal data. Nothing leaves the house.
- Knowledge chat
Chat with your own knowledge. Vector search plus a knowledge graph built by the local LLM. All models run locally.
- Team grows
More people? Add boxes. Rule of thumb for chat: one box per 4–15 people. New boxes register themselves.
Hundreds of support e-mails a day? A local workflow sorts each one: incident, request or noise.
- Master on a VM
Put the master on a VM. More CPU cores and RAM for Dify, Open WebUI, Docling, Cognee and the database. The boxes do the inference.
- Cloud: capacity or master
Peak load? Add EU cloud GPUs. The LLM Manager on your VM balances the load across boxes and cloud GPUs (S, M or L = 1, 2 or 4).
Or run the master in the cloud. The cloud instance takes over the control plane and orchestrates your boxes as workers.
- Developers
Add NVIDIA power. Your own RTX PRO 6000 server or a GB10 joins as a worker.
Developers connect their IDE. Any OpenAI-compatible client talks to llm.<domain>/v1 with its own rzfz-sk key.
- Coding agent 24×7
A coding agent that works 24×7. Agent Manager on the master: harness, local model and your Gitea. It writes tests and opens pull requests.
- Your own Mac
Already have a Mac Studio? Plug it in. Your hardware: supported, not sold by us. The LLM Manager serves its Ollama models.
- One fleet
Ours, yours, or both. One fleet. Every app talks to one door. The LLM Manager decides where each request runs.
Everything inside the red line runs in your own IT. Only the optional EU cloud runs outside.
- Outro
Local. Open source. Independent. Secure. Local AI on the hardware you choose.
Preinstalled & ready to run
Every bundle ships the rzfz.ai Stack prewired — including the Management UI you run it with: select modules, manage users, monitor status. No setup project, no waiting on integration.
Box + Stack
The turnkey appliance: hardware, stack and setup from one hand
- rzfz.ai Box (AMD Strix Halo, 128 GB)
- rzfz.ai Stack pre-installed
- Every box comes bundled with an active subscription. Year one included in the box price, then €799/box/year.
- Setup and day-1 handover
Cloud + Stack
Your dedicated European rzfz.ai cloud instance — handed over ready to operate
- Dedicated EU cloud instance (Small, Medium or Large — 96–384 GB VRAM)
- rzfz.ai Stack pre-installed
- Incl. subscription, update service, operations & support contract
Own hardware + Stack
You bring the servers, we bring the stack — mixed fleets welcome
- Installation on your Ubuntu 26.04 hosts
- Master VM + GPU worker topologies
- rzfz.ai Subscription licensed separately — €799/server/year
Mixed topologies
Many deployments combine bundles rather than committing to one: several boxes, cloud and your own hardware together, centrally managed through a shared control plane. Details on the individual profiles are under System requirements.
Get in touch