So they have decided that putting a small LLM on a phone would backfire because people would have a negative perception of their cloud models. Pretty sure AMD w

So they have decided that putting a small LLM on a phone would backfire because people would have a negative perception of their cloud models. Pretty sure AMD will use these taalas chips in data centers, not phones