Local hardware · No cloud

Legal AI that runs on a machine you can point at.

The whole system is a workstation in your office. You can see it, unplug it, lock it in a room, and know exactly where every matter it has read is sitting.

Request a Briefing See what it does

The specification, plainly

We recommend a machine. Firms buy, own and set it up themselves — there is no appliance to lease.

A current RTX-class GPU

24 GB of video memory minimum, 32 GB recommended. The GPU is what makes local inference on a whole matter practical rather than theoretical.

64 GB system memory, 2 TB SSD

Enough headroom for the corpus, the index and a large matter at the same time.

Windows, kept offline

A dedicated workstation rather than a shared server, designed to run with no internet connection.

Yours to administer

No appliance, no lease, no vendor remote access. It is a computer your firm owns, running software your firm bought.

What stays true on every matter

These do not change by practice area, firm size, or the question asked.

Nothing leaves the firmNo cloud, no API call, no telemetry. The matter is read on hardware the firm owns.

Owned, not subscribedA one-time purchase. No per-seat monthly fee and no usage meter.

Citations verified on-machineAuthority is checked against a local corpus, not recalled from a model’s memory.

The entire matterFull workup across the whole file — not a chat window over one document.

Runs on your workstationA current RTX-class GPU. Designed to run with no internet connection at all.

Get the specification before you buy anything.

We would rather you spec the machine correctly than discover the ceiling on your third matter. The briefing covers exactly what to buy.

Request a Briefing

Questions firms ask

The answers below are specific to this page’s subject.

Can it run on a laptop?

Not well. Whole-matter analysis is GPU-bound and needs the video memory. A dedicated workstation is the honest requirement.

Can we run it on a server we already have?

If it meets the specification, yes. The constraint is the GPU rather than the form factor.

What does run time actually look like?

It depends on the size of the case file and the hardware. Timing quoted anywhere on this site refers to a typical mid-sized matter on the recommended workstation; larger files take proportionally longer.

Do you sell the hardware?

No. We specify it and the firm buys it, which keeps ownership and administration unambiguous.

Other pages covering the same ground from a different angle.

On-premise legal AI platform Air-gapped legal AI software Offline legal AI for law firms Owned (not subscribed) legal AI Enterprise / large firm offline legal AI In-house legal team private AI

See it run on a real matter.

A briefing walks the whole path — intake, workup, the Adjudicator's pass, and the work product that comes out the other side.