On-premise. Preconfigured. Ready to use.

The plug-and-play AI solution for your company.

Local AI for your company knowledge – with preconfigured hardware, models and team access.

stark AI Box

Includes a knowledge base

Black, neutral workstation shown as an example stark AI Box

01

Inside your network

Data and models remain in infrastructure under your control.

02

Available offline

Local AI applications can operate without cloud access.

03

Prepared for teams

Accounts, permissions, APIs and requested models are configured initially.

04

Make knowledge available

Get answers from documents directly in the chat interface.

AI under your control

A system that goes beyond the hardware.

The AI Box combines capable workstation hardware with a ready-to-use software stack for secure, repeatable AI work in your company.

Unlock knowledge locally

Documents and internal knowledge can be made accessible through chat and RAG applications.

Control access cleanly

Team accounts, keys and permissions enable controlled use instead of shared personal accounts.

Integrate applications

Local APIs connect models with existing assistants and workflows.

Three performance tiers

The right box for your team.

Every box provides local AI with a knowledge base and team access. Starter is the cost-effective entry point; Box 32 and Box 48 add GPU memory, larger models and more headroom for concurrent use.1

For smaller companies

Box Starter

The simple entry point to local AI: secure team chat, internal knowledge search and compact assistants without cloud dependency.

For a few users and clearly defined tasks

Show technical details
BandwidthUp to 256 GB/s unified memory bandwidth
GPUAMD Radeon 8060S · integriert
CPUAMD Ryzen AI Max+ 395
RAM64 GB Unified Memory · LPDDR5X
Data1 TB NVMe
A fast start with local AI
Secure team knowledge search
Compact internal assistants

For everyday AI work

Box 32

For teams that need reliable research, content creation and internal knowledge assistants running in parallel.

4–8 users in parallel

Show technical details
BandwidthUp to 1,792 GB/s GPU memory bandwidth
GPUNVIDIA GeForce RTX 5090 · 32 GB VRAM
CPUAMD Ryzen 9 7900
RAM48 GB
Data1 TB NVMe
System2 × 512 GB SSD · RAID 1
Fast research and content creation
Internal knowledge assistants
Multiple workflows in parallel

For larger models and RAG

Box 48

For teams that need demanding analysis, large knowledge bases and complex automation securely inside their own network.

6–10 users in parallel

Show technical details
BandwidthUp to 1,344 GB/s GPU memory bandwidth
GPUNVIDIA RTX PRO 5000 Blackwell · 48 GB ECC-VRAM
CPUAMD EPYC 9654P
RAM64 GB
Data1 TB NVMe
System2 × 480 GB Enterprise SSD · RAID 1
Demanding analysis and automation
Large knowledge systems and RAG
Sustained use for larger teams

Included in the base package

Initial setup of team accounts

Setup of the requested models

Searchable knowledge base

Already configured

From the interface to governance.

Employees get an approachable chat interface. IT and accountable teams manage models, access, keys and usage rules through a local gateway.

Browser-based access

Use approved AI capabilities from the workplace without specialist software.

Team and permission management

Manage accounts and permissions transparently for groups and use cases.

APIs in the company network

Connect internal systems with approved models through controlled endpoints.

Monitoring and logs

Review operations, usage and technical events in one central place.

The software in detail

Open WebUI and LiteLLM.

Open WebUI provides the browser-based workspace. LiteLLM brings models and APIs together behind a controlled gateway.

Workspace

Open WebUI

A central browser entry point for chats, knowledge work and internal assistants inside the company network.

Gateway and governance

LiteLLM

A local access point for approved models, API keys, teams and traceable usage.

One box, two perspectives

Simple for employees. Controllable for IT.

01

Employees and knowledge work

A familiar interface for writing, summarising, researching and using internal assistants.

Central browser access

Approved models

Shared knowledge sources

02

IT and the company

A locally controlled entry point for models, integrations and transparent access rules.

Company network

Permissions and keys

Monitoring and governance

Build your own or choose a configured AI Box?

Both paths can make sense. The decision depends on your requirements, available IT capacity and responsibility for setup and operation.

  • Building your own offers flexibility but requires time for selection, setup, testing and ongoing maintenance.
  • For a configured system, the agreed scope and your team's involvement are clarified in advance.
  • Compare both paths using the same assumptions about usage, costs, security, updates and backups.
Build or buy an AI server? Read the comparison

For a custom integration of your knowledge sources: Plan a custom knowledge base

Introduction

From your use case to an operational AI Box.

The final configuration depends on models, user count, data sources and your existing infrastructure.

  1. 01

    Clarify requirements

    We frame tasks, user groups, data sources and integration needs.

  2. 02

    Configure the box

    Hardware and software are prepared for the agreed operating scope.

  3. 03

    Set up and hand over

    Accounts and models are configured initially and operations are handed over together.

Frequently asked questions

What to know before making an inquiry.

Can the AI Box be used without internet access?

Yes. The local AI applications are designed to run inside the company network without cloud access.

How many people can work at the same time?

As a guide, Box 32 is intended for about 4–8 and Box 48 for about 6–10 concurrently active users. Actual capacity depends on the model and tasks.

Can other models be used?

Requested models are configured initially. We assess which models run well on the selected hardware based on the use case.

Which documents can the knowledge base use?

Sources and access rights can be assigned individually.

Non-binding consultation

Which local AI setup fits your company?

We discuss your use case and help you choose the right configuration and setup. You do not need to select a box yet. We will get back to you by the next business day.

Your inquiry is not an order. If possible, describe your tasks, number of users, data sources and desired integrations; we can then clarify the appropriate configuration and scope.

team@stark-ai.de

Handled confidentially

Open privacy notice

Required fields are marked with *.