My outsider's understanding is Google is the only one whose custom silicon is the primary compute for their flagship foundation models. I didn't see any messagi

My outsider's understanding is Google is the only one whose custom silicon is the primary compute for their flagship foundation models. I didn't see any messaging about the Nova models being trained on Traininium (AWS), and Meta still talks about the number of H110's training their Llama models.