0
Applied AI·July 9, 2026·1 min read

PrismML says it ran a 27B-parameter Qwen 3.6 model on an iPhone 17 Pro, bigger than any prior on-device model; sources: Apple held talks with PrismML about it

Share

Running a 27B-parameter model on an iPhone 17 Pro is a shot across the bow for cloud-only AI — on-device is racing up the capability curve. If you’re designing AI products, revisit what must live server-side; latency, privacy, and cost math may flip faster than your roadmap assumes.