this post was submitted on 21 Jul 2026
207 points (100.0% liked)
PC Master Race
21819 readers
776 users here now
A community for PC Master Race.
Rules:
- No bigotry: Including racism, sexism, homophobia, transphobia, or xenophobia. Code of Conduct.
- Be respectful. Everyone should feel welcome here.
- No NSFW content.
- No Ads / Spamming.
- Be thoughtful and helpful: especially when new beginners have questions.
founded 3 years ago
MODERATORS
you are viewing a single comment's thread
view the rest of the comments
view the rest of the comments
I have an 8gb M1 Mac mini that my dad gave me. Put Asahi on it with a view to using it for my Home Assistant server. That on its own was perfectly fine, but adding an LLM for voice commands absolutely fucked the RAM. No amount of swap or anything could help, it'd just hang.
So I set it back to macOS, because it can hook into Apple APIs for LLMs that utilise the GPU in a way that Asahi couldn't. And yeah, Home Assistant + LLM runs really well even in a UTM VM. Same system, completely different outcomes.
And that's no shade to Asahi. What they've achieve is little short of incredible, but there's certain shit that Apple just won't let them access.