Comment by pdimitar

Oh, we can run LLMs efficiently with AMD GPUs now? Pretty cool, I haven't been following, thank you.