I do not believe this is correct. Sure you can run llama or something locally on a cpu or your Apple arm but it’s not even close to the speed of running on a Nv

I do not believe this is correct. Sure you can run llama or something locally on a cpu or your Apple arm but it’s not even close to the speed of running on a Nvidia card. Do you think nvidias trillion dollar valuation is because entire governments are buying Nvidia cards to run stable diffusion?!