[redleader55]

Local AI is almost impossible right now with the prices of RAM and GPUs and the sizes of decent models. No way spending even an optimistic 10k, but more likely 20k, on a setup that is good for 5-6 months makes any financial sense.

[bhhaskin]

Not yet, but it's only a matter of time. Local AI doesn't need to be bleeding edge either. Even capabilities of older modals would be fine.

[anakaine]

I disagree wholeheartedly. Older models do not perform anywhere near as well as newer models, and certainly not once you throw in agents that can sense check, security check, refa tor and balance, and research queries as they run behind the scenes.

[nozzlegear]

Just get a Mac?

[bassitone]

Impossible to find Mac minis in some areas, and if this goes through expect it to get worse.

I settled for the AMD rough equivalent. It’s not perfect but it can still handle most of the work. Now if only extra ram would come down in price… I find I need about 5 GB more than I have

[ares623]

Macs will just get more expensive. C'mon man, we're supposed to be Silicon Valley engineers, the pinnacle of human intellect.