--- title: "Framework Desktop review: the home AI machine you can actually open" canonical: "https://consideredgear.com/reviews/framework-desktop-395/" pubDate: "2026-09-07T00:00:00.000Z" author: Jim Shannon description: "128 GB of shared memory in a small box, running large AI models that a normal graphics card can't fit. It costs $1,999, and unlike most mini PCs you can take it apart." --- The Framework Desktop is a small computer built around AMD's Ryzen AI Max+ 395 chip, which comes paired with 128 GB of memory shared between the processor and a large built-in graphics chip. That shared pool is the whole point: it lets you load AI models that a $1,000 graphics card physically cannot hold. ## What the extra memory buys you Most people running local AI are working with 24–32 GB and mid-size models. This machine is for the tier above that — the large models that start to feel like the paid services. On 128 GB you can load one, keep it resident, and have it respond in a few seconds. It's not the fastest way to run a model that *does* fit in 24 GB — a dedicated graphics card wins that race. It's the only way to comfortably run the ones that don't. ## The Framework part Every other machine in this class is a sealed box. This one uses a standard mini-ITX case and standard parts. The storage is user-replaceable, the fans are standard, and Framework sells the mainboard on its own if you want to reuse the case later or move the guts into something else. The memory is the exception — it's soldered, because the chip requires it. So the 128 GB decision is permanent. Given that 128 GB is already the top option, that's easier to live with here than on a Mac mini. ## Who should buy it — and who shouldn't **Buy it if** you specifically want to run large models at home, you'd rather buy once than upgrade, and repairability matters to you. **Buy something else if** mid-size models are all you need — a [32 GB Mac mini](/reviews/mac-mini-m4/) does that for half the price and less power — or if speed on smaller models matters more than size, in which case a [used RTX 3090](/reviews/nvidia-rtx-3090-used/) in a PC you already own is the cheaper path.