I last tried out the beta version of it on Graphene a couple of months back and it was rough. In fairness though, I was mostly using it to figure out how to use it to install a full fat Linux distro, which is probably a niche use.
My phone's just updated, so I'll give it another spin when I get home tonight and report back.
The performance with regards to Home Assistant is spot on.
I have an M1 mini with 8Gb of RAM that I use as my Home Assistant server. I've given 4Gb to HAOS running in UTM, and let Ollama use the rest to run a modest LLM for speech to text. It's flawless.
I'm not suggesting that 8Gb is the gold standard or anything, but for some applications it's still perfectly usable.
I have an M2 Air with 16Gb, and for what I need it to do, I couldn't have any less, but the target market for a bargain basement entry level MacBook almost certainly won't ever notice. It'll literally just be a portal to access iCloud and whatever browser they use to get on their socials. And if they do find they need more, then Apple will happily sell them a new laptop.
Don't get me wrong, I think Apple's approach to RAM pricing has been criminal for years, and as a company they're really figuring out how to plumb the depths of scumminess, but on this I don't really see the issue.
With that said, the cost of an extra 8Gb to them is literally pennies, so withholding it is shitty. But it won't really make that much difference to the average performance of the computer.
I have an 8GB M1 mini in service as my Home Assistant server. 4GB to UTM to run HAOS, the rest for macOS and Ollama running a small LLM for speech to text. I'm genuinely amazed that it hasn't fallen over. Tried the same thing in Asahi but without macOS' memory management and access to GPU acceleration, it just wasn't feasible.