PrismML mete un modelo 27B en el iPhone. Esta versión conserva los hechos verificados y los ordena desde producto, infraestructura y gobernanza.
Qué ocurrió
PrismML mete un modelo 27B en el iPhone 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
Por qué importa
La señal de fondo es on-device AI is about cloud cost, privacy and latency as much as model size.
Qué observar
El próximo punto a comprobar es third-party reproduction on real iPhones and whether sustained inference matches the launch claim.
Fuentes：Sina Finance/Global Market Broadcast, PrismML site, GitHub, Hugging Face, The Information, AppleInsider, Techmeme, CocoLoop.