AI Local

Everything about running AI language models directly on your own machine or phone.

By Julian Harris

· Launched 2 months ago

Local AI starts behaving like a server

Reproducible runs, memory caps, and chunked prompts arrive.
May 20 • Julian Harris

AMD Wants the Local AI Crown. This is good for everyone.

The distant third horse in the race got a boost: AMD’s 128GB 200-billion-parameter-capable desktop is the beginning of a new dawn.
May 19 • Julian Harris

I said 1.4x was the limit. I was wrong.

2 days on and we’re seeing another 30% improvement in performance. What a wild ride.
May 16 • Julian Harris

Local AI by the gigabyte

Shopping lists by RAM tier, all-in-one products, and a Redis creator's 14-hour days
May 15 • Julian Harris

The Hardware Stood Still

Two years, one frozen memory ceiling, and a model that got nearly five times smarter anyway
May 14 • Julian Harris

Local AI on a Mac is actually working. With one giant catch.

The tokens-per-second numbers look great until your conversation resets and the laptop goes quiet for six minutes.
May 12 • Julian Harris

Local AI is starting to feel close

Five reports this week from the same hardware-first frontier, and one possible answer to "is it usable yet?" “Increasingly”.
May 10 • Julian Harris

A coding AI on your laptop, if you can afford the laptop

Three Apple Silicon enthusiasts say local AI for coding is here. But what spec are they talking about?
May 8 • Julian Harris

AI Local
Everything about running AI language models directly on your own machine or phone.