Google Tensor Processing Units were only introduced in May 2016. This truly a mind boggling breath taking rapid development!
It appears this is not truly a new chip, but it was introduced last summer, but apparently only to a selected group of users and for internal Google research.
"... In addition, TPU v4 has enabled breakthroughs at Google Research in the areas of language understanding, computer vision, speech recognition, and much more, including the recently announced Pathways Language Model (PaLM) trained across two TPU v4 Pods. ..."
No comments:
Post a Comment