Here’s why smartphones are getting more RAM

Apple unveiled its iPhone 16 lineup on Monday, and the big draw was Apple Intelligence. Apple’s AI system on the device offers flashy features like the ability to rewrite emails, generate custom emojis, and a significantly improved Siri. But beneath all of this, AI brings another big change to the iPhone: more RAM.

Although Apple never talks about RAM in its smartphones, Mac Rumors Uncovered that every iPhone 16 model now has 8GB of RAM, up from 6GB on last year’s base models. And it’s not just Apple making changes like that. Last month, Google made similar changes to its AI-powered Pixel 9; both the standard and pro models saw a bump in RAM, making 12GB the go-to option. least You can get it this year.

The impetus behind these RAM increases appears to be artificial intelligence. AI is the new must-have feature of the year, and it also consumes a lot of RAM. Smartphone makers are now increasing memory because they need to, whether they say it out loud or not.

AI models need to respond quickly when called upon by users, and the best way to accomplish this is to keep them permanently loaded into memory. RAM responds much faster than a device’s long-term storage; it would be annoying to have to wait for an AI model to load before you could get a quick summary via email. But AI models are also quite large. Even a “small” one, like Microsoft’s Phi-3 Mini, takes up 1.8 GB of spaceAnd that means taking away memory from other smartphone functions that previously used it.

You can see how this played out very directly on Pixel phones. Last year, Google didn’t enable local AI features on the standard Pixel 8 model due to “hardware limitations.” Spoiler: it was the RAM. Android VP and GM Seang Chau said In March that the Pixel 8 Pro could better handle Gemini Nano, the company’s smaller AI-powered model, because that phone had 4GB more RAM, at 12GB, than the Pixel 8. The model needed to remain loaded in memory at all times, and the implication was that the Pixel 8 would have wasted too much memory by supporting the feature by default.

“It wasn’t as easy to say, OK, we’re going to enable it on the Pixel 8 as well,” Chau said. Google eventually made Gemini Nano available on the Pixel 8, but only to people willing to use their phones in developer mode — people who Chau said “understand the potential impact on the user experience.”

Those trade-offs are why Google decided to increase RAM across the board with the Pixel 9. “We don’t want the rest of the phone experiences to slow down to accommodate the large model, so we increased the total RAM rather than sticking to the existing budget,” Google group product manager Stephanie Scott said in an email exchange with The edge.

So will all of that extra RAM go to AI alone, or will users see improved performance across the board? It’ll depend a lot on the implementation and how big those models are. Google, which added 4GB to support local AI features, says you’ll see improvements on both. “Speaking only of our latest Pixel phones,” Scott wrote, “you can expect better performance and improved AI experiences from your extra RAM.” He added that the Pixel 9 phones “will be able to keep up with future AI advancements.” But if those advancements mean larger models, that could easily mean they’ll eat up more RAM.

The same RAM-increasing trend is also playing out in the laptop world. Microsoft dictated earlier this year that only machines with at least 16GB of memory can be considered a Copilot Plus PC — that is, a laptop capable of running local Windows AI functions. rumored Apple is also planning to add more RAM to its next generation of laptops, after years of offering 8GB of RAM by default.

That extra memory will be necessary, especially if laptop makers want to keep even larger models loaded locally. “I think most operating systems will keep an LLM always loaded,” Julien Chaumond, Hugging Face’s chief technology officer, told me in an email, “so 6-8GB of RAM is the sweet spot that will unlock that in parallel with the other things the OS is already doing.” Chaumond added that models can load or unload “a small model on top of it to change some properties,” such as a style for image generation or domain-specific knowledge for an LLM. (Apple describes his approach similarly.)

Apple hasn’t explicitly said how much RAM is needed to run Apple Intelligence, but every Apple device that runs it, dating back to the 2020 M1 MacBook Air, has at least 8GB of RAM. Notably, last year’s iPhone 15 Pro, with 8GB of memory, can run Apple Intelligence, while the standard iPhone 15 with 6GB of RAM cannot.

John Giannandrea, Apple’s head of artificial intelligence he said in an interview in June Daring Fireball’s John Gruber said limitations like “device bandwidth” and the size of the neural engine would make AI features too slow to be useful on the iPhone 15. Apple’s vice president of software engineering, Craig Federighi, said during the same appearance that “RAM is one piece of the whole.”

The 2GB RAM increase on the iPhone 16 isn’t much, but Apple has been slow to expand the base RAM on its devices for a long time. Any increase in this area seems like a plus for usability, even if the company is starting small.

We don’t yet know how useful Apple Intelligence will be, or whether a small jump in memory will be enough to enable current iPhones to run the AI ​​features of the future. One thing seems certain, though: we’ll see more such hardware improvements as AI proliferates across the industry.