Free Shipping on all orders · Priority Mail Shipping with fee of $8.00
Now accepting engagements · Legal vertical

AI that reads your clients' documents.
Without the cloud.

AirGap AI runs a 70-billion-parameter language model on Apple Silicon hardware inside your firm. NDAs, contracts, privileged memos — analyzed on-device, with outbound network traffic physically disabled and verifiable at the OS level. Client data never leaves the machine.

What happens in the demo

A real NDA. Llama 3.3 70B. Wi-Fi physically off. lsof running live on screen to prove no network traffic. The model reads the document and flags unusual clauses — all on the laptop, with the receipts visible.

0

Outbound calls

Our code makes zero network connections. Audited every dependency. Verified live with lsof.

70B

Parameters on-device

Llama 3.3 70B runs at full 8-bit precision on Apple Silicon. Same class of model as cloud Claude.

$0

Per-query cost

No API fees. No per-seat pricing. Your hardware, your model, your data.

How it works

Three pieces, one MacBook.

1

Install

We set up a local AI server on your MacBook (M-series, 64 GB+ recommended). Double-click launcher. No command line for staff.

2

Train

We train two of your staff on the NDA + contract review workflow. 90-minute session, recorded for replay.

3

Review

Partners drop documents in, get structured clause analysis back. Wi-Fi can be off. Your client data never touches a third-party server.

Cloud AI vs AirGap AI

Same speed. Radically different trust model.

Scenario Cloud (ChatGPT, Claude Cloud) AirGap AI
Reviewing a sealed NDA ⚠️ Risky — data leaves the firm ✓ Stays on-device
Client asks about data retention ⚠️ Depends on provider ToS ✓ Nothing retained — no network call
Working on a plane / offline ✗ Doesn't work ✓ Works
Monthly cost for 4 partners $80–400+ recurring $0 after install
State Bar compliance burden ⚠️ Depends on jurisdiction ✓ Client data stays on-premise

Foundation engagement

A two-week fixed-scope deployment designed to establish AirGap AI inside your firm and document the workflow.

AirGap AI — Foundation

14 days · fixed scope · one firm per engagement. Pricing scoped during the intro call based on firm size and document throughput.
  • Deployment of the local AI stack on up to 3 attorney workstations
  • 90-minute training session for 2 staff members, recorded for team replay
  • Custom document-review workflow calibrated to your firm's contract and NDA templates
  • Written compliance memo suitable for your firm's legal records
  • 30 days of email support following handoff
  • Closeout report identifying high-leverage workflows and expansion opportunities
Request the intro call

Or email info@nicedreamzwholesale.com directly

Who this is for

We're narrow on purpose. Starting with law firms — the demo is built around NDAs and contract review — and expanding from there as the playbook matures.

Good fit:

Not the right fit yet (but tell us you exist, we're mapping the next verticals):

About Nice Dreamz

Nice Dreamz LLC designs and deploys local-first AI systems for firms that can't put client data on someone else's server. AirGap AI is our commercial offering for regulated industries — built on top of our open-source stack that now powers over 1,400 installations worldwide.

We publish our work in the open. Clients can audit every line of the inference layer, verify the on-device claims with their own tooling, and keep the stack running indefinitely whether we continue to support it or not. No lock-in, no black boxes, no license audits.

→ The open-source stack on GitHub

Honest questions we get

Is this as capable as ChatGPT or Claude Cloud?

For bounded professional tasks — NDA clause extraction, contract diffing, precedent comparison, document summarization — performance is comparable to cloud frontier models on the workflows we deploy. For open-ended reasoning or drafting complex briefs from scratch, cloud frontier models retain an advantage. Part of the engagement is mapping which tasks belong on-premise and which can safely route to cloud with appropriate redaction.

What does our IT team need to do?

Nothing outside standard application deployment. The stack runs as a local process listening on loopback only — no inbound firewall rules, no open ports, no VPN configuration changes. Installation is equivalent in complexity to deploying a standard Mac business application.

What hardware do we need?

Apple Silicon Macs (M2 / M3 / M4 / M5) with 64 GB unified memory or higher are the recommended configuration. For firms running 32 GB workstations, a smaller-parameter model provides good results for most document workflows. Hardware sizing is scoped during the intro call based on firm size and document volume.

What if the model produces an incorrect output?

AI outputs are always treated as first-pass material subject to attorney review — the deployment is structured around that assumption, not around replacing attorney judgment. Every engagement includes a written compliance memo documenting the tool's failure modes and the review workflow designed to catch them.

What happens after the engagement?

Your firm owns the deployed stack outright. Optional ongoing support covers new document types, quarterly model updates, additional workstations, and workflow tuning. Without ongoing support the system continues to operate indefinitely — no remote kill switch, no license check-in, no subscription dependency.

Why Nice Dreamz rather than a larger consultancy?

Larger consultancies typically quote six-figure engagements on multi-month timelines because their cost structure requires it. Nice Dreamz delivers the same outcome on a fixed-scope engagement because we built the underlying open-source stack and deploy it directly. Clients get the principal engineer for the full engagement, public source code they can audit without an NDA, and a clean exit path at the end of the pilot.

Schedule an introduction

A fifteen-minute conversation to understand your firm's document workflows, answer compliance questions, and determine whether the Foundation engagement is an appropriate next step.

Technical team prefers to evaluate the code first? The builders' community is on Discord: discord.gg/ZdSqgAxUW