Where it runs
Your model lives where you decide.
On your desk, in your building or on a dedicated GPU in your region. With Local, data never leaves. With Hybrid and Cloud, it stays in the region you choose.
Three homes
Eigna Local
On your servers, or an Eigna Box on your site.
Data never leaves.
Best for: banks, government and closed networks.
Eigna Hybrid
Small experts on your machine, the main expert on a GPU in your region.
Data stays in the region you choose.
Best for: teams whose machine is too small for the main expert.
Eigna Cloud
A dedicated GPU in the region you choose, run by us.
Data stays in the region you choose.
Best for: heavy token workloads and a fast start.
Where it runs
- Runs onEigna Box Business
- Data leavesNever
- KeysHeld by you
Compare the three
One model, where you decide.
| Compared | Eigna Local | Eigna Hybrid | Eigna Cloud |
|---|---|---|---|
| Runs on | Your servers or an Eigna Box on site | Small experts on site, the main expert on a GPU in your region | A dedicated GPU in the region you choose |
| Where data stays | Data never leaves. | Data stays in the region you choose. | Data stays in the region you choose. |
| Keys | Held by you | Held by you | Held by you, on your own volumes |
| Learning | Encrypted snapshot, rented GPU, deleted after the run | From approved work; path confirmed in your offer | From approved work; path confirmed in your offer |
| Best for | Banks, government, closed networks | Teams whose machine is too small | Heavy token workloads, a fast start |
| Setup | $3,250 a site | $3,250 a site | $5,500 a GPU |
| Running | $1,100 a site a month | $1,900 a site a month | $7,700 a GPU a month |
| Status | Planned | Planned | Planned |
On your hardware or ours
Your machines work too.
Apple silicon, NVIDIA RTX and GB10 machines with about 32 GB of memory or more. Or an Eigna Box: one base, five homes for the model, designed and built by us first.
Delivered and run
We deliver it, set it up and run it.
No ML team to hire. Your team names the task, approves the work and decides each release. We do the rest.
- 01UnderstandWeeks 1 to 3. The task, the data, the rules and today’s baseline.
- 02BuildWeeks 4 to 8. The home, the spec, and a sealed test the expert must beat.
- 03LiveFrom week 9. Installed, rolled out team by team, learning.
- 04CareA named contact, a monthly service report, and rollback at any time.
Timings are indicative until the first projects run. A new install from our templates targets under two weeks.
No seats. No token bill.
Setup once, then a fixed running cost per site or per GPU. Success should not make your AI more expensive.
Start
Not sure which mode fits?
The fit check recommends one in two minutes, from where your data must stay.