LokAI
Plug in. Power on. Prompt away.
Your device. Your office. Your data.
Powerful language models for your entire team. On a device in your office. LokAI stays where your data belongs. With you.
1ResearchAnswers in secondsFrom your documents or general knowledge2DocumentsDrafts, reports, summariesCreated from your templates3KnowledgeYour internal know-how becomes part of the modelTemplates, workflows, guidelines — all available4Maintenance & UpdatesNew models, security updates, supportWe take care of it — you work
LokAI_local intelligence
One device.
Your entire team.
LokAI makes your team more productive.
For those who want to get more from their knowledge.
Instant
Real-time responses
< 10 Min
Unbox to first prompt
Unlimited
No caps, no recurring fees
Legal
Healthcare
Enterprises
Developers
What happens in your office,
stays in your office.
Pricing
The right box for your team.
Handcrafted in AustriaShips in 4–6 weeks
Entry
Small teams & law firms
VRAM
24 GB
Performance
46.9 TFLOPS
CUDA Cores
8,960
Token Speed
~50 token/s
20B model, batch size 1
- RAM64 GB ECC
- Storage1 TB NVMe
- Modelsmodels up to ~20B parameters
- Usersup to 5 concurrent users
- Operation24/7 operation
from €5,000
4× VRAM · 4× Token/s
Pro
Large teams & intensive workloads
VRAM
96 GB
Performance
125 TFLOPS
CUDA Cores
24,064
Token Speed
~200 token/s
20B model, batch size 1
- RAM64 GB ECC
- Storage1 TB NVMe
- Modelsmodels up to ~120B parameters
- Usersup to 20 concurrent users
- Operation24/7 operation
from €15,000
Custom
Enterprise & data center
VRAM
Configurable
Performance
Configurable
CUDA Cores
Configurable
Token Speed
Configurable
Depends on configuration
- RAMAs needed
- StorageAs needed
- ModelsUnlimited
- GPUsMulti-GPU
on request