Comment by andy99

Running local LLMs split across machines. Pretty sure the author has some other posts about doing that.

Edit: https://www.jeffgeerling.com/blog/2025/15-tb-vram-on-mac-stu...