REDACTEDAI SOLUTIONSDiscuss deployment

Pricing + buying process

Know the starting point before we ask for a meeting.

Private AI is a considered purchase. The price should still be visible up front. Start with one bounded workflow, a visible service fee, hardware selected for the work, and a written path from evaluation to production.

Request a pilot or quote ↗See exactly how it works ↓
$7,500Bounded starting scope
45 daysDefined evaluation
≈80–120 hrsTypical delivery effort
SixConcrete outputs

Why $7,500

The fee buys a bounded implementation, and you can see where the hours go.

Your fee funds a founder-led implementation that turns one sensitive workflow into a working local evaluation, tests whether it deserves to scale, and leaves your organization with evidence and documentation.

STARTING PROJECT FEE$7,500
funds
TYPICAL DELIVERY EFFORT≈80–120 hours
to produce
DECISION-READY OUTPUTSystem + evidence
WHERE THE WORK GOESAPPROXIMATE
01

Boundary + architecture

Workflow discovery, users, approved data, authority, facility constraints, hardware profile, risk, and success measures.

12–18 hrs
02

Grace + R.A.I.S.E. environment

Local runtime, orchestration, access boundaries, policy gates, records, and the secure operating path.

18–28 hrs
03

Workflow + intelligence

Model selection, retrieval, source preparation, tools, prompts, and one usable primary workflow.

18–28 hrs
04

Governance + evaluation

Known-answer tests, failure review, human checkpoints, authority limits, calibration, and user feedback.

18–28 hrs
05

Documentation + handoff

Findings, limitations, operating guidance, meetings, support, and a written production recommendation.

14–18 hrs
WHAT LEAVES WITH YOU06 OUTPUTS

A system decision your organization can defend.

  • A working local private-AI evaluation environment
  • One configured, governed workflow using approved sources
  • Documented tests, failures, limitations, and review points
  • A hardware and infrastructure profile based on observed demand
  • Operating documentation your organization can retain
  • A scale, revise, or stop recommendation, with no forced continuation
The value math:

At a typical 80–120-hour starting scope, $7,500 implies roughly $63–$94 per delivery hour before overhead. This is not hourly billing. Your proposal defines the price, boundary, and outputs so the engagement cannot drift into an open-ended consulting project.

What the starting price includes

The pilot fee buys evidence inside a defined boundary.

The pilot stays narrow by design. Its purpose is to answer expensive architecture, governance, hardware, and workflow questions before your organization commits to a larger production investment.

INITIAL PILOT BUDGET$7,500 service+approved hardware+special licenses, if any
INCLUDED IN THE STARTING SCOPE08
  • Founder-led workflow and boundary design
  • One approved primary workflow and source collection
  • Grace OS orchestration and governance configuration
  • R.A.I.S.E. local environment configuration
  • Local model and retrieval path selected for the workload
  • Policy gates, review points, and human accountability
  • Representative testing and documented failure findings
  • Operating documentation and a scale, revise, or stop recommendation
QUOTED SEPARATELY WHEN REQUIRED06
  • Workload-fit edge, workstation, server, or cluster hardware
  • Unusual third-party model, software, or data licenses
  • Large-scale document remediation, scanning, or migration
  • Custom application interfaces and broad enterprise integrations
  • Formal certification, authorization, or regulated-compliance programs
  • Production redundancy, multi-site rollout, and continuing support

Three ways to engage

Start where your organization is today.

01 / BEST FIRST STEP

Private AI pilot

From $7,500+ hardware

One bounded 45-day workflow designed to establish value, controls, usability, failure behavior, and the right production profile.

  • Known starting price
  • Defined end date
  • No forced continuation
Start the pilot path ↗
02 / CUSTOM SCOPE

Production deployment

Written quoteper environment, office, or node

For organizations ready to deploy Grace OS and R.A.I.S.E. across a defined environment, including hardware, installation, acceptance, documentation, and handoff.

  • Edge to GPU-cluster scale
  • Local or air-gapped
  • Deployment-specific controls
Request a production scope →
03 / CONTINUING OPERATIONS

Support + evolution

Ongoing planquoted after system scope

Continuing support for the approved environment, including controlled change, evaluation, documentation, recovery review, and priority technical assistance.

  • Model and dependency changes
  • Evaluation and configuration
  • Operational support
Discuss a support plan →

The buying process

From first click to working local system.

No hidden checkout, no sensitive-data upload, and no obligation to scale after the pilot. You see the boundary, price, hardware, responsibilities, and deliverables in writing before work begins.

01

Choose a path

Use the pricing page or inquiry form to identify whether you need a pilot, a production deployment, or ongoing support. Do not send sensitive material.

02

Founder fit call

A no-charge first conversation confirms the workflow, reason for local operation, internal owner, timing, and whether Redacted AI Solutions is an appropriate fit.

03

Boundary + technical scope

We define users, approved data, tools, agent authority, model needs, facility constraints, success measures, and the consequence of failure.

04

Written proposal

You receive the exact service fee, hardware profile, licensing assumptions, schedule, responsibilities, exclusions, and deliverables before committing.

05

Authorize the engagement

The start date is reserved after both parties approve the written agreement and the required initial invoice is paid. No sensitive data enters before the boundary is approved.

06

Build + evaluate

For a pilot, we configure the local system, run the bounded workflow, test known questions and failure behavior, collect user feedback, and document the results across 45 days.

07

Scale, revise, or stop

The pilot ends with evidence. Continue into a per-environment, per-office, or per-node production proposal only when the value and operating fit justify it.

WHAT CHANGES THE PRICE

Model scale, context size, users, concurrency, corpus condition, integrations, agent authority, facility and isolation requirements, hardware, uptime, regulated obligations, travel, and support expectations. We identify these variables before the written quote. They do not show up for the first time on an invoice.

Direct answers

Questions worth asking before you sign.

What does “starting at $7,500” mean?+

It is the service starting point for one tightly bounded 45-day private-AI pilot. A larger corpus, multiple workflows, complex integrations, regulated requirements, on-site work, unusual licensing, or more extensive engineering will change the written quote.

Why does the pilot start at $7,500?+

This is not a software subscription or access fee. A typical starting scope is designed around approximately 80–120 hours of founder-led discovery, architecture, Grace and R.A.I.S.E. configuration, workflow engineering, governance, evaluation, documentation, and handoff. At the starting scope, that works out to a blended delivery rate of roughly $63–$94 per hour before business overhead. The written scope and outcomes set the price. The engagement is never billed as an open-ended hourly project.

Why is hardware priced separately?+

The right hardware depends on the model, context size, users, concurrency, storage, uptime, facility, power, cooling, portability, and isolation requirements. Separating hardware keeps a small workflow from subsidizing a cluster, and keeps a demanding workflow off an undersized box.

Can we go directly to production?+

Yes, when the organization already has mature requirements, an approved use case, an internal owner, and a realistic technical and security process. A bounded discovery and written architecture scope still come first. For most new workflows, the pilot is the lower-risk path.

Is this an online checkout product?+

No. Private and air-gapped AI affects data handling, hardware, authority, operations, and security. The buying path stays direct: fit call, boundary, written proposal, agreement, invoice, and scheduled delivery.

What does ongoing support cover?+

Support is scoped after the deployment. It can include controlled model and dependency updates, evaluation, configuration changes, backup and recovery review, documentation, incident coordination, and priority technical support. The exact plan and fee depend on the environment.

Can we send documents for a quote?+

Do not send confidential, privileged, regulated, export-controlled, patient-identifiable, or otherwise sensitive material through the public website. The first scope uses descriptions and sanitized examples; an approved transfer path is established later when required.

Ready when the workflow is real

Start with a no-charge fit call and leave with a clear next step.

Choose “Pricing, pilot, or production quote” on the inquiry form. Ralph answers personally and helps determine whether the next move is a pilot, a production scope, more preparation, or no engagement at all.