Services AI Audit Guides Marketplace Blog Contact
Local AI

Hardware for local models. Memory is the whole story.

One rule explains most of this: the model must fit in memory. Everything else — speed, model choice, whether you need a GPU — follows from that constraint. Get memory right and the rest is optimisation.

Sizing, roughly

A quantised model needs approximately its parameter count in gigabytes, plus overhead for context. An 8-billion-parameter model at common quantisation wants around 8–10GB. A 70B model wants 40GB or more.

Memory availableComfortable model sizeWhat that gets you
8GB3B, small 7B quantisedBasic summarising and classifying. Usable, limited.
16GB7B – 8B comfortablyThe practical entry point. Good for most business tasks.
32GB13B – 14B, 30B quantisedNoticeably better quality. A good default target.
64GB+70B quantisedClosest to cloud quality available locally.

Unified memory changed the picture

Apple Silicon Macs share memory between processor and graphics, so a machine with 64GB of unified memory can load a model that would otherwise require an expensive dedicated GPU. For a small business wanting one capable machine without building a workstation, a Mac with generous memory is often the simplest good answer — and it is quiet enough to sit in an office.

On the PC side, video memory on the graphics card is the binding constraint rather than system memory. A card with 24GB of VRAM handles considerably more than one with 8GB, regardless of how much system RAM the machine has.

Does it need a GPU?

For usable speed, effectively yes — or unified memory doing the same job. Models run on CPU alone, but generation slows to the point where people stop using the system, which is the real failure mode. A tool that takes forty seconds to answer gets abandoned no matter how private it is.

Configurations that make sense

Trying it out

The laptop you already own, if it has 16GB. Costs nothing, and tells you within a week which workflows local AI can actually handle.

One person, daily use

A Mac with 32–64GB unified memory, or a PC with a 16–24GB graphics card. This is the sweet spot for most small businesses.

Small team

A dedicated machine with a strong GPU on the network, serving several people, with authentication in front of it.

No hardware purchase

A rented GPU instance. Monthly cost, but a private instance rather than a shared service — a reasonable middle ground for regulated data.

Buy for the model you will actually run daily, not the largest one you might experiment with. A fast 8B model used constantly delivers more value than a 70B model that is technically impressive and too slow to fit into anyone's working rhythm.

Common questions

Can I run a local LLM on a normal laptop?

With 16GB of memory, yes — an 8B-class model runs comfortably and handles summarising, drafting and classification well. With 8GB you are limited to small models and will feel it. Below that, local inference is not a realistic proposition.

Do I need an expensive graphics card?

You need either a capable GPU or Apple Silicon unified memory. CPU-only inference works but is slow enough that people abandon the tool, which defeats the purpose. The cheapest viable path is usually a Mac with generous unified memory rather than building a PC around a high-VRAM card.

Is it cheaper than paying for cloud AI?

It depends entirely on volume. For occasional use, no — hardware costs far more than a few dollars of API calls. For continuous high-volume work, hardware pays back within months and then costs nothing per use. Estimate your monthly API spend honestly before buying anything.

What about running models on a rented server?

A good middle path: a private GPU instance gives you a dedicated environment without capital expenditure, and for regulated data a single-tenant instance is often acceptable where a shared consumer service is not. You pay monthly, but the instance is yours rather than a shared service.

Keep reading
Ollama for business Local vs cloud, compared Running AI locally

AI that runs
on your terms.

Private AI systems in environments you control — your data never trains public models, and engagements are available under NDA.

Private AI systems Book a free audit