Apple in Early Talks With Startup on Compressed AI Models for iPhones

Apple is in talks with PrismML, a Silicon Valley startup, about technology that compresses AI models to run on iPhones. PrismML publicly released compressed versions of Alibaba's open-source Qwen model on Tuesday. The company reduced the model from roughly 54 GB to less than 4 GB, allowing its 27 billion parameters to run on an iPhone 15 or newer.

PrismML CEO Babak Hassibi said Apple and other companies are evaluating the startup's models for speed, energy efficiency and performance on devices. "They're really evaluating our technology right now," Hassibi said of Apple. " The release came one day after Apple opened the public beta of iOS 27.

PrismML shrinks models by reducing each internal value from 16 bits to one or three possible values. The compressed models use 10 to 15 times less memory, generate responses 6 to 8 times faster and consume 3 to 6 times less energy than conventional versions, according to the company.

Hassibi said the models typically lose a few percentage points of overall performance compared with uncompressed versions.

The technology emerged from Hassibi's research group at Caltech, which owns the underlying patents and licenses them exclusively to PrismML. In March, PrismML raised a $16.25 million seed round backed by Khosla Ventures and other investors.

Hassibi said Google's open-source Gemma model is next in the pipeline. Morgan Stanley estimates Apple's average DRAM cost per bit could rise roughly 190% year-over-year in fiscal 2027, with NAND costs up about 180%.