Apple is reportedly evaluating PrismML, a startup specializing in compressing AI models to enable their operation directly on iPhones, according to CNBC. PrismML has successfully reduced Alibaba’s Qwen model from around 54GB to under 4GB, allowing it to run on iPhone 15 devices and newer, with improvements in memory usage, response speed, and energy consumption. This move aligns with a broader trend among major tech firms focusing on on-device AI processing to enhance user privacy and optimize performance in consumer electronics.

Apple: Apple Inc. is a major technology company that develops consumer electronics, software, and services centered around devices like the iPhone. In the context of this news, Apple is reportedly assessing PrismML’s model compression capabilities to support running advanced AI directly on its hardware.
PrismML: PrismML is a startup developing techniques to compress large AI models for efficient operation on edge devices such as smartphones. The company is being evaluated by Apple for its approach to enabling powerful models to function on iPhones with lower resource demands.

`json
{
“On-Device AI Trend”: “Major tech firms are focusing on enabling artificial intelligence processing directly on mobile devices to enhance user privacy and improve response times.”,
“Model Optimization Focus”: “Startups in the AI industry are developing techniques to compress AI models to fit consumer electronic hardware while maintaining a balance between performance and hardware resource requirements.”
}
`