Apple previously clarified that the Neural Engine’s capabilities determine if it can run generative AI features. In short, the company has hinted that the unit in the iPhone 15 Pro and iPhone 15 Pro Max can handle the taxing nature of Apple Intelligence, which only means that the upcoming A18 running in the iPhone 16 family can run it well, but by how much? Based on one rumor, the new chipset will be more powerful than the M4, which Apple claims is the fastest Neural Engine in any of the firm’s chipsets.
The M4 running in the latest 11-inch and 13-inch iPad Pro models is currently the most powerful and efficient Apple Silicon found in a device bearing an incredible thinness level, with the technology giant stating that the chipset also features the fastest Neural Engine, operating at 38 trillion operations per second. Based on this claim, Apple states that the M4 is faster than any NPU found in current-generation AI PCs, and according to @MappleGold on X, the California-based giant is expected to kick it up a notch with the A18, which is scheduled to debut later this year.
We expect the A18 and A18 Pro to be mass produced on TSMC’s 3nm ‘N3E’ process, and while this advanced lithography is essential to maintain that impressive ‘performance per watt’ ratio, the Neural Engine will ultimately decide how well the chipset can handle generative AI features. The exact performance metrics were not shared on the micro-blogging social network, but assuming that the A18 is more powerful than the M4 in this specific regard, we could see Apple include a figure higher than 38 trillion in its press release. However, there is another thing that the rumor mentions.
The company may take a
Read more on wccftech.com