[saurik]

A big part of this device seems to be trying to provide the memory bandwidth needed for high-performance local LLM inference, in a package that is more designed to be controlled by the user than the Apple Mac Mini/Studio that have so far been the best for such work.

[thisislife2]

I doubt if Apple CPUs / GPUs outperform AMD CPUs and GPUs but yes, there have been a lot of these "AI" mini-PCs suddenly mushrooming everywhere - https://www.gmktec.com/products/amd-ryzen™-ai-max-395-evo-x2... ...

[ktallett]

I question the term best fit. Do you mean best fit within an off the shelf pre built device?