Step 01
1. Plug in LionBox
Connect power and Ethernet. The mobile app finds the nearby box and starts guided setup.
From first setup
How does LionBox work?
LionBox combines a guided mobile setup, a browser-based AI workspace and dedicated local GPU compute. Your team sends requests over the office network; LionBox processes them inside the appliance and returns each answer privately.
Your browser
“
Supplier agreement.pdf
Local request
Inside LionBox
100% localCompany knowledge
Local AI model
Your answer
Local response
Answer ready, with sources
Setup for non-technical teams
From sealed box to first prompt in minutes.
Three guided steps. No specialist deployment and no software to install on employee laptops.
Step 02
2. Configure using the mobile app
Claim the box, connect the network, choose models and create accounts for your team.
Step 03
3. Open the workspace
Authorized users open LionBox in a local browser and start working—nothing to install.
Inside every request
What happens after you press Send.
The browser, permissions, knowledge search, model and response path all stay within your office network and LionBox hardware.
01
Local browser
A signed-in user sends text, an image or a document from the LionBox interface.
02
Access check
LionBox applies the user’s permissions before any private or shared knowledge is opened.
03
Context assembly
Chat history, memory and relevant RAG sources are prepared for the selected model.
04
Local GPU inference
The request is scheduled on the dedicated NVIDIA GPU with other active team requests.
05
Private answer
The answer returns to that user over the LAN. The prompt and response never need the cloud.
No external inference APINo cloud prompt historyPermission-aware knowledgeLAN-only response path
Local performance architecture
One GPU. Multiple private workstreams.
A dedicated high-capacity NVIDIA GPU runs the selected model. LionBox schedules concurrent requests across available compute, keeps user sessions separate and assembles the right context for each answer.
Dedicated NVIDIA GPU
Local inference is accelerated on appliance hardware instead of a remote API.
ABC
01
02
03
Concurrent request scheduler
Active team requests are queued and scheduled against available GPU capacity.
Saved locally
Chat history and memory
Conversations and user memory stay attached to the correct private workspace.
Integrated local RAG
Relevant company and private-user documents are retrieved with their permissions intact.
LionBox mobile app
One app to run the whole box.
The administrator app guides the technical work in plain language. Configure, monitor and update LionBox without turning everyday operations into an IT project.
Selected view
Network & models
LionBox V1
Network & models
Office network
LionBox SecureConnected
Installed models
Qwen3.5 9B
Gemma 4 12B
Ministral 3 14B
Add a model
App video placeholder
First-time setup walkthrough
Claiming the box, connecting the network and installing the first model.
App video placeholder
Everyday administration
Users, permissions, health checks, alerts, working hours and model updates.
Local browser workspace
Your team’s daily AI workspace.
Chat, attach files, search approved knowledge and revisit source-backed answers in the browser—without installing desktop software.
LionBox Control
System healthy
LionBox Assistant
I'm running locally on LionBox. Ask me about your company knowledge base, or try the sample question below.
Local model · No external API · Demo data only
Available AI models
Choose the model that fits the work.
Choose a tested model in the mobile app. Each option is ready to run on LionBox for the kinds of work shown below.
AvailableQwen3.5 9B
Multimodal model
- Inputs
- Text + images
AvailableGemma 4 12B IT
Multimodal instruction model
- Inputs
- Text + images
AvailableMinistral 3 14B Instruct
General instruction model
- Inputs
- Text + documents
See more models
More model profiles are coming soon.
New models will appear in the mobile app after they are prepared and tested for LionBox hardware.
Measured on LionBox V1
Speed you feel. Headroom you can trust.
More than 2,000 benchmark messages across realistic office work, simultaneous users, surge traffic and long documents.
0
failed requests
No errors, dropped accepted requests or memory failures across more than 2,000 benchmark messages.
0.07 s
Typical wait before the first word
The interface starts responding in under a tenth of a second during everyday use.
~2,000words/min
Delivered to each active user
Around eight times faster than average human reading speed.
<0.5 s
Wait with 16 simultaneous sends
Even when 16 people press Send together, every reply begins in under half a second.
5.2 s
To read a 90-page report
LionBox processed roughly 23,000 words and returned an answer in just over five seconds.
Local intelligence, taking shape
Your AI. Your advantage.
Claim your spot in the future of private AI.
Reserve your Founders Edition for €5 + VAT. Save €500 at launch. Cancel before shipping.
Keep me in the loop
Be first to hear.
Get the launch announcement and delivery updates in your inbox.
Founders Edition · 50-unit allocationA little deposit.
A little deposit.
A €500 head start.
€5 + VAT · fully refundable deposit
Your €5 + VAT deposit locks in the €1,999 price. The remaining €1,994 is due only when your hardware is ready to ship. The software license is €29 per month, billed quarterly from activation. Taxes and shipping will be confirmed separately. Secure deposit checkout powered by Revolut.
Made in Europe
LionBox V1 · Founders Edition
€5 + VAT deposit · €500 launch discount