[ad_1] Google works with Broadcom to help her design AI AI AI chips, known as tensioner processing units (TPU). Keep in mind that these are not the same as the GX tensioner processors used to operate the pixels. According to a new report, Google may replace Broadcom as a Google design partner in TPU. The report says that the designer of MediaTek will intervene to work on the new TPU, which will be the seventh generation AI chips. To reduce their dependence on NVIDIA graphics units, which are the most widely used chips used to train artificial intelligence models, Google designed TPU AI. Which is allocated to the use of artificial intelligence and is used by Google for internal work burdens. It is also used by Google Cloud customers. As a result, Google does not rely on NVIDIA as much as the main artificial intelligence players. Google's competitors like Openai and Meta are still highly dependent on NVIDIA, which can bring in reverse results when there is deficiency. Google TPU for automated learning. | Credit-Google image Graphics processing units are used in graphics processing units to help provide graphics and images on devices such as smartphones, and they are known for their ability to process large amounts of data simultaneously. This is more in line with the matrix pattern to process the data used with artificial intelligence. Central processing units are more known for their serial style in data processing, making it less useful for Amnesty International.
[ad_2]
Download
Google to get surprising new partner to help design its seventh-gen AI chip
Despite this report, it should be noted Google does not cut its relationships with Broadcom although there are strong reasons for Google to select Mediaatek. One of the reasons is that MediaTek has a great relationship with Taiwan -based TSMC, which is the largest chip -paced pair in the world that will allow Mediaatek to impose a less fee on Google for each segment of the amount that Broadcom receives. Last year, Google was believed to have spent between $ 6 billion and 9 billion dollars on TPUS, according to omdia.
For example, Open AI Sam Altman CEO said at the end of last month His company has run out of the NVIDIA GPU that forced Openai to release the new GPT-4.5. I asked at least 15 times a day Although I have never been asked why AI models are used as GPU foil instead of the CPU chips, I will give you the answer.
| Name | |
|---|---|
| Publisher | |
| Genre | News & Magazines |
| Version | |
| Update | March 17, 2025 |
| Get it On |  |
