You could also buy the Apple Studio with its large amount of unified ram for a similar price of a 5090. Of course it’s not as fast but it could run a model that needs more ram.
barebones 780m amd mini pc. 128gb ddr5600. About $700. Under $800. Vulkan is good, non AMD supported version of ROCm works for some, because AMD says fuck you to its users. $2000 higher end mini pc with 8600s gpu is other option for high ram with under double the performance, but closer for cheaper than apple solution.
The pricing for memory is still pretty bad. $4K for 96GB, $5.6K for 256GB, $10K for 512GB. One can get 128GB on the M4 Max for $3.5K, at the cost of a narrower bus so it’s even slower, but generally, EPYC + a 3090 or 4090 makes a lot more sense.
SOTA quantization for these are mostly DIY. There aren’t many MLX DWQs or trellis-quantized GGUFs floating around.
But if you want to finetune or tinker instead of just run, you’re at an enormous disadvantage there. AMD’s Strix Halo boards are way more compatible, but not standalone yet and kinda rare at this point.
You could also buy the Apple Studio with its large amount of unified ram for a similar price of a 5090. Of course it’s not as fast but it could run a model that needs more ram.
barebones 780m amd mini pc. 128gb ddr5600. About $700. Under $800. Vulkan is good, non AMD supported version of ROCm works for some, because AMD says fuck you to its users. $2000 higher end mini pc with 8600s gpu is other option for high ram with under double the performance, but closer for cheaper than apple solution.
The pricing for memory is still pretty bad. $4K for 96GB, $5.6K for 256GB, $10K for 512GB. One can get 128GB on the M4 Max for $3.5K, at the cost of a narrower bus so it’s even slower, but generally, EPYC + a 3090 or 4090 makes a lot more sense.
SOTA quantization for these are mostly DIY. There aren’t many MLX DWQs or trellis-quantized GGUFs floating around.
But if you want to finetune or tinker instead of just run, you’re at an enormous disadvantage there. AMD’s Strix Halo boards are way more compatible, but not standalone yet and kinda rare at this point.