Bringing PyTorch Monarch to AMD GPUs

Love monarch, amazing primitives and feels so much lighter than Ray.

Another awesome add on their part

bicepjai 4 hours ago | parent | prev | next [-]

So are we at a point where we can’t train llms for fun at home on less expensive AMD cards ?

esseph 3 hours ago | parent | next [-]

Who was training LLMs on residential AMD GPUs?

	▲	MrDrMcCoy 35 minutes ago \| parent [-]
		I am. Most models that can be run locally have extraordinarily out of date training data for certain frameworks that predate breaking changes and important features, so their docs and API schemas need to be put into LoRAs.

amelius 4 hours ago | parent | prev [-]

can't or can?

fefe23 2 hours ago | parent | prev [-]

benchmqarks?