Law office bookshelf server: Mac mini M4 Pro 48GB + AnythingLLM per matter
A local legal stack guide: a silent Mac mini M4 Pro with 48GB unified memory running Ollama with one AnythingLLM workspace per matter, plus encryption and a written AI policy.
Solo2-10 peopleDocument search / chat with filesWriting assistant$1k-3kSensitivity: Very high (legal/medical/financial)
Hardware
| Item | Approx. price (as of 2026-10-16) |
|---|---|
| On sale −6% MSI GeForce RTX 5060 Ti 16GB Ventus 3X OC (Newegg, related current offer) | $579.99 |
| On sale −5% Apple Mac mini M4 Pro (12C CPU/16C GPU, 48GB, 512GB) (B&H, related current offer) | $2,099.00 |
| Apple Mac mini (M4 Pro, 48GB unified memory) | $2,200-2,500 (as stated by source) |
Source prices are as stated by the source. Retailer offers: prices checked 2026-10-16, may change. Some links may be affiliate links; we may earn a commission at no extra cost to you. This never changes what we list.
Software stack
- Ollama
- AnythingLLM
- FileVault full-disk encryption
Models used
- Llama 3.3 70B (Q4) - source claims 8-15 tok/s
Good at
- Silent, always-on, 30-50W idle (per source)
- Workspace per matter, delete on closure
- LLM, embeddings and vector DB fit in unified memory
Weak at
- Slower than a 24GB NVIDIA GPU (per source)
Rough total cost
~$2,200-2,500 hardware (source figure, retrieved 2026-10-16).
Getting started
- Buy/configure the Mac mini with 48GB and enable FileVault.
- Install Ollama and pull the chosen model.
- Install AnythingLLM; create one workspace per matter.
- Ingest PDFs and require citations in answers.
- Write a one-page AI usage policy; delete workspaces when matters close.
Source & credit
Local AI for lawyers (May 2026) — runlocalai.co
Summarized from the source; quotes are the source author's words. Check the original before buying.