Or if you have a printer/scanner combo, you can turn it into a pen pal!

How about the iMac Pro? Would that work? I was able to put 128gb in it (not as easy as the regular iMac but possible).

I've been running various models on a Mac Pro 2013 (8 cores, 32 GB RAM) at about 8 to 10 t/s for months. It's not fast, but it's more than enough for many actual tasks, in particular background tasks. An iMac pro will do just as well I suppose.

I have and use a Mac Pro 2013 too. Mine is 8 cores with 64 GB RAM. I haven't used mine for any LLM workloads, but it does just fine for most stuff. My biggest concern with it is the OS. I'm still running macOS (the latest supported version) but it's getting continually further out-of-date security wise all the time.

What are the tasks that do well with 8-10 t/s ?

The sort of task you don't expect to end immediately. If extracting data from a bunch of PDFs takes 1 hour or the whole night, that doesn't make much difference to me. It's not fast enough for auto completion and slightly too slow for chat (but bearable IMO).

Running a local llm at 10 t/s overnight to extract data from a few PDFs will burn more in electricity than paying cents for the hosted kimi models.

You can (sometimes) break even if you have a workstation GPU.

Sometimes data privacy is paramount.

Responsibility, reputation, reliability, human connection and empathy (ahem lawyers, right…)

Maybe just the ones at full or hybrid war with Europe?

The landscape has changed significantly. In 2007, OS X itself (10.5 Leopard) cost $129.

Maybe just ai-gapped.

Is that an offhanded joke on the terminology or do you actually mean something? I can't tell.

Been doing it since the day I was born. The beginnings were hard but I’m getting there.

You've actually been primarily training a physics model, with an LLM attached to it.

Good point, and I'm actually not sure that there is a clear dividing line. I expect that once we achieve capable world models and are able to analyze their internals, we'll find that the prediction mechanisms for purely physical and for verbal/behavioral responses to the agent's actions are at least partially colocated.

As particular motivation for my intuition, I expect that we had evolutionary pressure to adapt our defense mechanisms of predicting the movements of predators and prey, to handle human opponents.

Happy to help and eventually take over.

I recommend the recent Wookash podcast with Chuck Jazdzewski, who was part of the team that created the original TV and much more in that ecosystem.

Interesting, looks like it's two completely separate implementations, one in Swift and one in Python.

It is exactly that. The macOS and GNOME versions share the same vision, but they are entirely different codebases.