[cultofmetatron]

can these do training or only inference? currently working on learning machine learning and I'd love to have a physical machine I could aim to build real workloads on in a few years.

[LogicFailsMe]

They're Turing complete. What else do you need?

[Our_Benefactors]

It’s possible (likely, even) to have a chip fast enough for inference, but not fast enough or with enough memory to do meaningful training runs. Like the current DGX spark.

[Npovview]

There is a reason why Google has tpu8i and tpu8t

[porphyra]

technically in order for something to be turing complete it needs infinite memory

[LogicFailsMe]

The more (memory) you buy, the more you save!

[the_real_cher]

I believe training is way more processor intensive than inference.

[airjason]

not for llm full training, but can do some finetuning for sure.