July 7, 2024

News Collective

Complete New Zealand News World

Apple Intelligence will come to the iPhone 15 Pro due to the combination of the chip and RAM

Apple Intelligence will come to the iPhone 15 Pro due to the combination of the chip and RAM

Madrid, June 20 (Portaltic/EP) –

the Apple intelligence features It will not reach all Apple devices because not all of them are equipped internally to give the support that large language model implementations need, although it will reach the iPhone 15 due to its chip and RAM.

Apple Intelligence, Apple’s artificial intelligence system, which will be integrated with it iOS 18, iPadOS 18, and macOS Sequoia To access iPhone, iPad and Mac with a differentiated approach: it combines generative AI with each user’s personal context to deliver a meaningful experience.

Although it is expected to arrive in the next devices of the technology company, it will also do so in some models that have already been marketed, as is the case with iPhone 15 Pro series, But not the iPhone 15.

This is because Combination of processor and RAM used by more advanced iPhones, due to the computing demands of large language models, as conveyed by company executives after WWDC, and now includes MacRumors.

specially, A17 Pro chip It contains a neural engine or neural processing unit, consisting of 16 cores with the ability to perform 35 billion operations per second. A must have for all compatible devices Random access memory (RAM) of at least 8 GB.

“This is the hardware it takes. It’s pretty extraordinary to run models this powerful on an iPhone,” said Craig Federighi, head of software engineering.

“These models, when run at runtimes, are called heuristics, and Inference from large language models is incredibly computationally expensivesaid Head of Artificial Intelligence and Machine Learning, John Gianandrea.

See also  The real reasons behind the death of Google Stadia have emerged

This manager also explained, “This is a combination of the bandwidth on the device, which is the size of Apple’s neural engine, which is the engine on the device to speed up these models enough to be useful.”