Apple Eyes Startup That Squeezes AI Models Small Enough for an iPhone

Apple is in early talks with a small Silicon Valley startup that claims it can shrink powerful artificial intelligence models to run directly on an iPhone, the company's CEO told CNBC. PrismML, a Caltech spinout backed by Khosla Ventures, released compressed versions of Alibaba's open-source Qwen model on Tuesday, slashing its size from roughly 54 GB to under 4 GB. That lets all 27 billion of the model's parameters operate on an iPhone 15 or newer. PrismML CEO Babak Hassibi said Apple and other companies are evaluating the technology, measuring its speed, energy efficiency and performance on devices. He described the discussions as very early but noted "things are progressing nicely."

The release comes a day after Apple opened the public beta of iOS 27, giving users their first broad access to the company's long-delayed Siri overhaul. Apple is trying to make Siri more competitive with assistants from OpenAI and Anthropic while keeping more personal data and AI processing on the device. Running more AI directly on the iPhone would reduce delays, lower cloud costs and support Apple's privacy pitch. It would also let features work without an internet connection. Analyst Carolina Milanesi of Creative Strategies said smaller models could allow Apple to move demanding features like computational photography, video generation and health tools onto the phone. "The more you can do on device, the better it is," she said.

PrismML shrinks models by simplifying how internal information is stored, reducing each value from 16 bits to just one or three possible values. The company says this cuts memory use by 10 to 15 times, speeds up responses by six to eight times and slashes energy consumption by three to six times compared to conventional versions. Hassibi acknowledged a trade-off: the compressed models lose a few percentage points of performance, particularly in factual recall. The startup is releasing two compressed versions for free, designed to run on iPhones, MacBooks and Nvidia-powered PCs. The technology emerged from Hassibi's research group at Caltech, which owns the underlying patents and licenses them exclusively to PrismML. In March, the company raised $16.25 million in seed funding from Khosla Ventures and others.

Source: CNBC