Apple is in talks with the startup shrinking a 27B AI model onto an iPhone

Apple is in early talks with PrismML, a startup that compresses large AI models to run on iPhones. PrismML's Bonsai 27B model, shrunk from 54GB to 3.9GB, can fit on an iPhone 17 Pro, potentially enabling more on-device Siri processing. The move could reduce cloud costs, improve privacy, and help Apple manage rising memory expenses.

Apple is in early talks with PrismML, a startup that shrinks large AI models to run directly on a phone. The company’s on-device AI pitch could help Apple keep more of Siri’s work off the cloud. PrismML chief executive Babak Hassibi told CNBC https://www.cnbc.com/2026/07/14/apple-prismml-ai-compression-iphone.html that Apple and other companies were evaluating its technology. He called the discussions very early and said it was unclear where they would lead, but that “things are progressing nicely.” Apple did not comment. The Information first reported Apple’s interest last week. The startup is a Khosla Ventures-backed spinout from the California Institute of Technology. Caltech owns the underlying patents and licenses them exclusively to PrismML. The company raised a $16.25 million seed round in March. What PrismML built On Tuesday, PrismML released Bonsai 27B https://prismml.com/news/bonsai-27b . It is a compressed build of Alibaba’s open-source Qwen https://thenextweb.com/news/alibaba-qwen-apple-intelligence-china-approved model, not a new one trained from scratch. The company shrank it from roughly 54GB to as little as 3.9GB. PrismML ships two versions under a free licence. A ternary build runs on a laptop. A smaller 1-bit build, about 3.9GB, is designed to fit within the memory budget of an iPhone 17 Pro. PrismML says it is the first model of that size to run on a phone. The trick is how the model stores its internal values. PrismML reduces each one from 16 bits to just one or three possible values. It says this cuts memory use by 10 to 15 times, speeds up responses by six to eight times, and lowers energy use by three to six times. There is a cost. Hassibi said the compressed models lose a few percentage points of performance. Factual recall weakens first, he said, before skills such as reasoning, maths, and coding. PrismML says its builds keep about 95% of full performance in the ternary version and 90% in the 1-bit one. Why Apple cares The timing is not an accident. PrismML released the model a day after Apple opened the public beta of iOS 27 https://thenextweb.com/news/apple-wwdc-2026-siri-ai-gemini-ios-27 , which carries its long-delayed Siri overhaul. Apple is trying to make Siri competitive with assistants from OpenAI and Anthropic. Running more AI on the device would help. Apple already sends complex requests to cloud models https://thenextweb.com/news/apple-siri-google-gemini-nvidia-privacy-wwdc . Keeping more work local would cut delay, lower cloud costs, and support the company’s privacy pitch. Some features would also work offline. There is a cost angle too. Morgan Stanley estimates Apple’s memory costs could climb sharply in its 2027 financial year. The bank expects the company to raise iPhone prices https://thenextweb.com/news/apple-to-raise-prices-as-memory-chip-shortage-bites-tim-cook-says to protect margins. Smaller models help Apple fit capable AI into tight hardware without paying for more memory. Claims still to be proven Analysts urged caution. Tarun Pathak of Counterpoint Research said the real test would be millions of queries across thousands of devices. Phil Solis of IDC said power use was the biggest open question, since a model that runs often could still drain a battery. The release also feeds a debate over whether efficiency gains will cut demand for memory and data-centre chips. Gil Luria, an analyst at D.A. Davidson, said shrinking models would not remove the need for processors. It would simply move some of them from data centres onto phones, part of a broader shift toward edge AI https://thenextweb.com/news/syntiant-us-ipo-edge-ai-nasdaq . Hassibi said Google’s open-source Gemma model is next in the pipeline, followed by larger frontier models. “It’s very important that the intelligence be local and that it can run fast,” he said. Get the TNW newsletter Get the most important tech news in your inbox each week.