May god have mercy on my power bills… the local AI cluster grows 😍
Mac Studio Ultra 256GB joining the fleet, time to get familiar with the Mac side of local AI. Anyone have tips or tricks for getting up to speed quickly?
Everytime I see the "why spend money on hardware for local AI, just use subscriptions" argument, I die a little on the inside.
These two tweets cover the two absolutely critical pieces that often get missed in this "debate".
Privacy abuses by providers get worse, while your limits shrink.
Don't be a subscription-slave, choose freedom tech via local AI.
@MiaAI_lab I’d guess the Sparks will still win on prefill and concurrency, but not by much…
Tempted to return the 4x we just got and get 2x 256GB Ultras instead 🤔
Thoughts?
Be forewarned if you’re using OCEAN for mining that you’ll be automatically forced into signaling for the BIP 110 fork if you don’t change your stratum endpoint in a week 🫠
The irony of forcing their miners to signal for 110 and abusing their centralized power as pool operators is off the charts 🤦♂️