llama_drama on Microsoft researchers developed a hyper-efficient AI model that can run on CPUs · Hackerly

Comment on Microsoft researchers developed a hyper-efficient AI model that can run on CPUs parent

I wonder if instructions like VPTERNLOGQ would help speed these up