TapTechNews readers Mr. Hangkong and Mr. Wuyingge provided clues! TapTechNews reported on May 15th that Google announced the sixth-generation data center AI chip Tensor Processing Unit (TPU) - Trillium at the I/O 2024 developer conference today, and said it will be delivered later this year. Google CEO Sundar Pichai said, Google was born for this moment. For over a decade, we have been pioneers in GPU technology. Pichai stated that after the new TPU is delivered, the computing performance of each chip will be 4.7 times higher than the fifth generation, achieved by expanding the matrix multiplication unit (MXU) of the chip and increasing the overall clock speed. In addition, Google has doubled the memory bandwidth of the Trillium chip. Trillium uses the third generation SparseCore, which Google describes as a dedicated accelerator for common large-scale tasks in high-ranking and recommendation workloads. The company believes that this will allow Trillium TPU to train models more quickly and provide services to models with lower latency. Pichai also described the new chip as Google's most energy-efficient TPU to date, which is particularly important as the demand for artificial intelligence chips continues to grow exponentially. Pichai's translation of some content is as follows: In the past six years, the industry's demand for ML computing has increased by a million times, roughly 10 times a year. If we don't invest in reducing the power consumption of these chips, this situation will be unsustainable. Google promises that the new TPU is 67% more energy-efficient than the fifth-generation chip.