Skip to content
AI & Local Models

AI & Local Models

Models that run on your own hardware rather than someone else’s — an LLM server, and two dictation apps that transcribe locally.

The common thread is that inference happens locally. Apple Silicon’s unified memory makes a Mac a credible inference host, which is the same argument as self-hosting anything else — no per-token cost, no rate limit, and nothing leaves the machine.

This post is licensed under CC BY 4.0 by the author.
Last updated on