▲ 336 ▼ OpenAI Hoarding Tens Of Thousands Of Apple Mac mini And Mac Studio Devices, As ASUS And MSI Burn Through Their Entire First Batch Of NVIDIA RTX Spark Chip And Beg For More. (wccftech.com) submitted 1 month ago by eicker@lemmy.world to c/technology@lemmy.world 79 comments fedilink hide all child comments
[–] melroy@kbin.melroy.org 5 points 1 month ago (5 children) Just wait for Mac m5 ultra permalink fedilink source parent hideshow 5 child comments replies: [–] eicker@lemmy.world [S] 13 points 1 month ago (4 children) The Mac Studio already has it. And from October, it’ll also be available with 512GB. It does cost around 18K, though. permalink fedilink source parent hideshow 4 child comments replies: [–] melroy@kbin.melroy.org 6 points 1 month ago (1 child) M5 ultra as indeed up to 512gb of unified memory. So you can run very large models. At the same time it's much much faster memory speed and bandwidth, almost simular to the nvidia rtx pro 6000. Sure the price is also high. But you get much better value for this setup compared to what nvidia is delivering you today. I still want to see even more competition.. Not just from Apple. permalink fedilink source parent hideshow 1 child comment replies: [–] xSikes@feddit.online 1 point 1 month ago What I’ve been reading around lately, that’s too much for the daily/major of users. permalink fedilink source parent [–] xSikes@feddit.online 1 point 1 month ago (1 child) That’s pretty crazy. I’ve been reading that models around 14-32B is plenty for daily tasks and then some. permalink fedilink source parent hideshow 1 child comment replies: [–] eicker@lemmy.world [S] 1 point 1 month ago The bigger the model the bigger the memory requirements. permalink fedilink source parent
[–] eicker@lemmy.world [S] 13 points 1 month ago (4 children) The Mac Studio already has it. And from October, it’ll also be available with 512GB. It does cost around 18K, though. permalink fedilink source parent hideshow 4 child comments replies: [–] melroy@kbin.melroy.org 6 points 1 month ago (1 child) M5 ultra as indeed up to 512gb of unified memory. So you can run very large models. At the same time it's much much faster memory speed and bandwidth, almost simular to the nvidia rtx pro 6000. Sure the price is also high. But you get much better value for this setup compared to what nvidia is delivering you today. I still want to see even more competition.. Not just from Apple. permalink fedilink source parent hideshow 1 child comment replies: [–] xSikes@feddit.online 1 point 1 month ago What I’ve been reading around lately, that’s too much for the daily/major of users. permalink fedilink source parent [–] xSikes@feddit.online 1 point 1 month ago (1 child) That’s pretty crazy. I’ve been reading that models around 14-32B is plenty for daily tasks and then some. permalink fedilink source parent hideshow 1 child comment replies: [–] eicker@lemmy.world [S] 1 point 1 month ago The bigger the model the bigger the memory requirements. permalink fedilink source parent
[–] melroy@kbin.melroy.org 6 points 1 month ago (1 child) M5 ultra as indeed up to 512gb of unified memory. So you can run very large models. At the same time it's much much faster memory speed and bandwidth, almost simular to the nvidia rtx pro 6000. Sure the price is also high. But you get much better value for this setup compared to what nvidia is delivering you today. I still want to see even more competition.. Not just from Apple. permalink fedilink source parent hideshow 1 child comment replies: [–] xSikes@feddit.online 1 point 1 month ago What I’ve been reading around lately, that’s too much for the daily/major of users. permalink fedilink source parent
[–] xSikes@feddit.online 1 point 1 month ago What I’ve been reading around lately, that’s too much for the daily/major of users. permalink fedilink source parent
[–] xSikes@feddit.online 1 point 1 month ago (1 child) That’s pretty crazy. I’ve been reading that models around 14-32B is plenty for daily tasks and then some. permalink fedilink source parent hideshow 1 child comment replies: [–] eicker@lemmy.world [S] 1 point 1 month ago The bigger the model the bigger the memory requirements. permalink fedilink source parent
[–] eicker@lemmy.world [S] 1 point 1 month ago The bigger the model the bigger the memory requirements. permalink fedilink source parent