TPU v5p

Google's TPU v5p and upcoming Ironwood TPU7x offer higher raw matrix throughput and simpler interconnects than NVIDIA H100 GPUs for large-scale AI training, but GPUs remain superior for inference with…

Researchers at UC San Diego, led by Hao Zhang, have successfully implemented a diffusion-style speculative decoding method called DFlash on Google TPUs, achieving an average 3.13x increase in tokens p…