Local AI pricing

How we usually price a Local AI deployment: hardware, one-time implementation, and ongoing support. Final numbers depend on scope, but this page shows the structure.

Baseline 3‑year scenario

Example for a 5‑person team using a Mac Mini Local AI stack—numbers from the Local AI overview page:

Item Local AI (Mac Mini) Cloud subscription
Hardware purchase $1,599 (one‑time) $0
Implementation labour $2,500 (one‑time) $0
Monthly subscription $0 $150 ($30 / user / mo)
Total cost (36 months) $4,099 $5,400

These numbers are examples, not a formal quote. After we’ve scoped your situation, we put together an offer that matches your team size and constraints.

Example packages

Minimal (1–3 seats)

Solo / small pilot

One Local AI stack for an individual or micro‑team that wants to test workflows before a bigger rollout.

  • Hardware recommendation and base Mac Mini setup.
  • Ollama + 1–2 Llama models installed and configured.
  • Open WebUI + a simple Obsidian vault connected to RAG.
  • 1–2 priority use cases configured end‑to‑end.
  • One focused training session.

Typical range (excluding hardware): $2,000–$3,500.

Best when you want to validate Local AI on real work with a minimal but complete setup.

Mid-size (5–10 seats)

Small team rollout

One Mac Mini Local AI stack for a small team—good for leadership, legal, finance, or an innovation group.

  • Hardware recommendation and Mac Mini setup.
  • Ollama + Llama models installed and configured.
  • Open WebUI, Obsidian, and RAG engine connected.
  • Up to 3–4 priority use cases configured end‑to‑end.
  • Team training session + Q&A.

Typical range (excluding hardware): $4,000–$6,000.

Good for teams that want Local AI in daily work without a full department‑wide change programme.

Big (10–25+ seats)

Department / multi‑team rollout

Designed for a full department (e.g. legal, finance, operations) or multiple teams that need Local AI embedded in daily work.

  • High‑availability Local AI host(s) sized for your load.
  • Role‑based access and workspace structure in Open WebUI.
  • Department‑specific Obsidian vault(s) and plugin setup.
  • 5–8 use cases configured with prompts, dashboards, and automations.
  • Team training sessions, office hours, and rollout support.

Typical range (excluding hardware): $8,000–$15,000.

Suited for organisations that want measurable impact, governance, and clear ownership of the Local AI stack.

Custom package

Tailored stack & automations

For clients that need something beyond the standard tiers—multiple locations, strict compliance, or deep integrations with existing systems. We treat the Minimal, Mid-size, and Big models as starting points and then shape a custom packet around your organisation.

  • Joint scoping of data sources, tools, and security requirements.
  • Custom infrastructure design (on‑prem, hybrid, or mixed vendors).
  • Advanced automations, dashboards, and reporting.
  • Change‑management and training plan matched to your organisation.

Pricing is fully custom and based on the agreed scope. We put together a written offer after scoping, so you know exactly what will be delivered.

Discuss a custom Local AI package

Hourly services

For smaller engagements or follow‑up work, services are billed at $300 per hour. Common items:

Obsidian Core Setup
2 hrs · $600
Smart Connections Config
4 hrs · $1,200
Copilot & Ollama Linking
2 hrs · $600
Custom Prompt Engineering
3 hrs · $900
Vector Indexing Optimization
4 hrs · $1,200
Staff Training (per team)
2 hrs · $600

Ongoing maintenance is available at the same rate. Many clients book a 2‑hour monthly slot for model updates, RAG tuning, and new use‑case experiments.

How we create your offer

Pricing always starts with a scoping conversation: we map out your data sources, security requirements, and 3–10 use cases. After that you get a written offer so you know exactly what we’ll build and what it will cost.

Talk about Local AI pricing