Apple Makes Major Breakthrough in AI Research, Enabling Complex AI Systems to Run Smoothly on iPhones
-
According to VB, Apple recently published two papers showcasing significant advancements in its artificial intelligence research. One of the new technologies could potentially allow complex AI systems to run smoothly on devices with limited memory like iPhones and iPads.
In the papers, Apple researchers addressed a key challenge: how to deploy large language models (LLMs) on devices with limited memory.
Apple's engineers have minimized data transfer from flash to memory through a series of methods, improving inference latency by 4-5 times and achieving 20-25 times acceleration on GPUs. This breakthrough is particularly important for deploying advanced LLMs in resource-constrained environments, greatly enhancing applicability and accessibility.
For Apple users, these optimizations may soon allow complex AI assistants and chatbots to run smoothly on iPhones, iPads, and other mobile devices.
Previously, Ming-Chi Kuo, an analyst at TF International Securities, stated in a recent report that the iPhone16 will introduce AI-related innovative features. It is reported that Apple reorganized its Siri team in the third quarter of this year with the goal of integrating AIGC functionality and LLM (Large Language Model).