PrismML, 27B 모델을 iPhone에 탑재. 확인된 사실을 유지하면서 제품, 인프라, 거버넌스 관점에서 현지 독자가 읽기 쉽게 정리했다.
무슨 일이 있었나
PrismML, 27B 모델을 iPhone에 탑재 is the current news peg. The verified record centers on these points:
- PrismML says it compressed a 27B dense Qwen 3.6 model to under 4GB for local iPhone 17 Pro use
- the company says an open model will be released next Tuesday
- Bonsai work lists 1-bit and 1.58-bit compression with local CPU and device backends
- real tests must check latency, heat, KV cache memory and quality loss
왜 중요한가
더 큰 신호는 on-device AI is about cloud cost, privacy and latency as much as model size.
다음 관전점
다음 확인 지점은 third-party reproduction on real iPhones and whether sustained inference matches the launch claim.
참고 자료：Sina Finance/Global Market Broadcast, PrismML site, GitHub, Hugging Face, The Information, AppleInsider, Techmeme, CocoLoop.