● Covered by 2 sources · 2 reportsMedium impact
Apple Engages PrismML for On-Device AI Model Compression
Key points
- Apple in talks with PrismML for AI model compression on iPhones.
- PrismML compresses AI models to run on-device with 27 billion parameters.
- The Qwen 3.6 model can operate on iPhone without server support.
- Discussions are in early stages; impacts on performance and privacy noted.
Apple and PrismML Discuss AI Model Compression
Apple is reportedly in discussions with PrismML, a startup known for its ability to compress large AI models for on-device use on iPhones. This could enhance the device's AI capabilities without requiring server support.
PrismML's technology allows significant models like the Qwen 3.6, containing 27 billion parameters, to run on iPhones. This offers potential increases in performance and privacy for users.
Potential Improvements for On-Device AI
PrismML has managed to shrink the Qwen 3.6 model from roughly 54 GB to less than 4 GB. This substantial reduction allows the model to run fully on devices such as the iPhone 15 or newer, providing them with advanced processing capabilities.
Having these capabilities on the device can minimize the need for external server computation, which could not only increase efficiency but also strengthen user privacy.
Early Stages of Negotiation
CEO Babak Hassibi of PrismML shared that Apple, along with others, is in the preliminary phases of evaluating PrismML's technology. Tests focus on the speed, energy efficiency, and performance of the model when operated on-device.
While the outcomes of these discussions remain pending, they signify Apple's efforts to bolster its on-device AI functionalities.
Broader Implications for Apple's AI Strategy
The discussions align with Apple's broader strategy to enhance device capabilities, reflected by its ongoing innovations in software and recent AI acquisitions. As Apple continues to refine Siri and other AI-driven features, technologies like PrismML's could become pivotal.
The impact on the tech industry hinges on whether Apple integrates such technologies into mainstream products, potentially setting a new standard for on-device AI processing.
✨ This summary was generated by AI from the outlets' reporting listed below. It is not independently verified and may contain errors — check the original sources. How BrevFeed works →
The daily brief
One email each morning: the day's tech stories, clustered across outlets and summarized. No account needed.
One email a day. Unsubscribe in one click, any time.
Today's brief
Spend a few minutes, get the whole day. Every topic's top stories in one hands-free rundown — listen, watch, or read the transcript.
▶ Play today's brief
New every morning, and the back catalogue is archived by date.
How outlets covered it
Apple is negotiating with PrismML, a startup that can compress AI models to operate on iPhones. This development may facilitate advanced on-device AI capabilities in iPhones, enhancing performance and privacy.
Apple is reportedly interested in PrismML, a startup that enables advanced AI models to operate on iPhones without servers. The technology, which compresses large models like Qwen 3.6, could enhance on-device AI capabilities for Apple devices.