I would pickup an inference GPU to run models locally. I can see a benefit in that especially if they’re on the cheap

Read his multipart on arguing with AI boosters. He covers silly arguments like this.

Also to paraphrase Cory Doctorow, you’re not going to keep breeding these mares to run faster and then one day they’ll birth a locomotive…

Spicy perhaps?

Does this mean thunderbolt works yet? Only reason I cant switch to Asahi on M1.

Thank God I bought the GOTY on sale years ago and it looks great with a few tweaks and mods…

This is me looking at myself cause I use both lol