Apple researchers recently published two papers describing techniques for using LLM technology in limited-memory environments. Chatbots based on LLM (Large Language Model), such as ChatGPT, require a huge amount of memory to function. This is why a lot of research is taking place to use this technology on smartphones, including the iPhone, with limited memory capacities.
To solve this problem, Apple researchers have developed a new technology that uses flash memory to store data from AI models. In a paper titled “LLM in Flash: Efficient LLM Inference with Limited Memory,” Apple explains that flash memory is more efficient in mobile devices than DRAM, which is traditionally used to run an LLM.
By integrating an LLM into the iPhone, the company is able to use data that has already been processed instead of loading new data each time. The use of flash memory allows you to reduce data transfer by reusing part of the data already processed.
Towards a smarter Siri
Taken together, these methods can run models up to twice the size of DRAM, four to five times faster than CPU speed, and 20 to 25 times faster than GPU speed, the researchers said. researchers.
Apple is said to have organized an internal event on AI next February to inform employees about working at home on LLMs. And Bloomberg reports that Apple is developing a smarter Siri that incorporates generative AI technology.
Apple is also working on an in-house AI model called “Ajax.” Designed to compete with OpenAI’s GPT-3 and GPT-4 models, Ajax aims to unify machine learning development within Apple. Ajax would be more advanced than ChatGPT 3.5, but OpenAI’s latest model, GPT-4, would still be ahead in terms of performance.
Apple is expected to introduce generative AI on its iPhones and iPads when it releases iOS18 according to analyst Jeff Fu of Haitong Securities and Bloomberg. Jeff Fu said last October that Apple had deployed hundreds of AI servers this year and would deploy more next year.
Apple tends to avoid using buzzwords like “AI” to describe the features of its products, preferring to focus on machine learning. However, these research papers suggest a deeper engagement with new AI technologies. However, Apple has not publicly acknowledged integrating generative AI into its products and has yet to officially confirm its work with Apple GPT.
Source: “ZDNet Korea” and “ZDNet.com”