TechOfficeApple Launches Its Inaugural Local AI Event
Apple's Inaugural Local AI Event: A Glimpse into the Future of Memory and On-Device AI
On June 23rd, Apple Park played host to a groundbreaking event as Apple conducted its first official local AI event, signaling a pivotal shift in the tech giant's focus toward on-device artificial intelligence. This exclusive gathering featured high-profile executives, including CEO Tim Cook and hardware chief John Ternus, who delivered invaluable insights regarding the development of local AI technologies.
A Strong Statement on Memory Configuration Roadmaps
According to tech analyst Alex Cheema, the event was not only about discussing local AI, but it also hinted at Apple's commitment to high-memory configurations amidst ongoing DRAM shortages. Several slides presented during the event alluded to robust memory settings of 512GB and 2TB, raising expectations around Apple's next-generation chips.
Cheema speculates that Apple’s upcoming M5 Ultra chip will likely be launched with a unified memory of 512GB, while its rumored successor, the M7 Ultra, is anticipated to house up to 1.5TB of memory, effectively maintaining Apple's reputation for high-performance devices.
Unpacking the Memory Capabilities for AI Inference
Although Cheema was unable to present the original slides, he reconstructively captured the details of a particularly noteworthy slide titled "Enabling a Full Range of Local AI Inference." The summary is as follows:
| Device | Unified Memory | Active Parameters |
|---|---|---|
| iPhone and iPad | Up to 16GB | Up to 14 billion |
| MacBook Air | Up to 32GB | Up to 35 billion |
| Mac mini | Up to 64GB | Up to 70 billion |
| MacBook Pro | Up to 128GB | Up to 120 billion |
| Mac Studio | Up to 512GB | Up to 480 billion |
| Mac Studio Cluster | Up to 2TB | 1.6 trillion+ |
Future Outlook: New Chip Developments
This update does not only encapsulate immediate advancements. Bloomberg's renowned journalist, Mark Gurman, corroborated these insights, revealing that Apple is laying the groundwork for an AI server built upon the M5 Ultra chip, codenamed J246. Additionally, plans are in motion for a second-generation server chip based on the forthcoming M7 Ultra, expected around 2029. Further out, Apple is also focusing on an M8 chip (codenamed "Soko") and a high-end Mac chip, referred to as "Cardinal," both anticipated to be manufactured on a cutting-edge 1.4nm process by 2028.
Conclusion: A New Chapter in Apple's Technology Landscape
Although Apple has not officially confirmed any of these developments, the implications of this event suggest a robust trajectory towards enhancing on-device AI capabilities. As the landscape of technology continues to evolve, Apple’s commitment to high-memory configurations and local AI solutions demonstrates its strategic foresight and determination to remain at the forefront of innovation in the tech industry.
The swell of excitement generated from this inaugural local AI event portends a transformative future for Apple's product ecosystem, as the company strategizes to harness the full potential of artificial intelligence across its devices.
👨🏫 Apple held its first-ever official local AI event which is quietly confirmed its memory roadmap. 📰 According to Alex Cheema, Apple hosted a private local AI event at Apple Park on June 23rd — Apple's first event of its kind, with Tim Cook, John Ternus and other execs speaking about on-device AI. 😐 Cheema says several slides referenced 512GB and 2TB memory configurations – a strong signal Apple isn't planning to abandon high-memory options despite the ongoing DRAM shortage. He expects the M5 Ultra to ship with 512GB unified memory, with the rumored M7 Ultra reaching up to 1.5TB. 🧠 He wasn't able to share the original slides, but recreated one from memory –"Enabling a full range of local AI inference": - iPhone and iPad. Up to 16GB unified memory. Up to 14 billion active parameters. - MacBook Air. Up to 32GB unified memory. Up to 35 billion active parameters. - Mac mini. Up to 64GB unified memory. Up to 70 billion active parameters. - MacBook Pro. Up to 128GB unified memory. Up to 120 billion active parameters. - Mac Studio. Up to 512GB unified memory. Up to 480 billion active parameters. - Mac Studio Cluster. Up to 2TB unified memory. 1.6 trillion+ active parameters. 🤔 This lines up with Bloomberg's Mark Gurman, who separately reported Apple is preparing an AI server based on the M5 Ultra (codename J246), a second-gen server chip based on the M7 Ultra for ~2029, and – further out – an M8 chip (codename "Soko") plus a high-end Mac chip ("Cardinal"), both on a 1.4nm process by 2028. P.S. None of this is officially confirmed by Apple. @Apples 👨🏫 Apple held its first-ever official local AI event which is quietly confirmed its memory roadmap. 📰 According to Alex Cheema, Apple hosted a private local AI event at Apple Park on June 23rd — Apple's first event of its kind, with Tim Cook, John Ternus and other execs speaking about on-device AI. 😐 Cheema says several slides referenced 512GB and 2TB memory configurations – a strong signal Apple isn't planning to abandon high-memory options despite the ongoing DRAM shortage. He expects the M5 Ultra to ship with 512GB unified memory, with the rumored M7 Ultra reaching up to 1.5TB. 🧠 He wasn't able to share the original slides, but recreated one from memory –"Enabling a full range of local AI inference": - iPhone and iPad. Up to 16GB unified memory. Up to 14 billion active parameters. - MacBook Air. Up to 32GB unified memory. Up to 35 billion active parameters. - Mac mini. Up to 64GB unified memory. Up to 70 billion active parameters. - MacBook Pro. Up to 128GB unified memory. Up to 120 billion active parameters. - Mac Studio. Up to 512GB unified memory. Up to 480 billion active parameters. - Mac Studio Cluster. Up to 2TB unified memory. 1.6 trillion+ active parameters. 🤔 This lines up with Bloomberg's Mark Gurman, who separately reported Apple is preparing an AI server based on the M5 Ultra (codename J246), a second-gen server chip based on the M7 Ultra for ~2029, and – further out – an M8 chip (codename "Soko") plus a high-end Mac chip ("Cardinal"), both on a 1.4nm process by 2028. P.S. None of this is officially confirmed by Apple. @Apples
