TechOfficeApple Unveils Its Inaugural Local AI Event, Setting a New Benchmark in Innovation
Apple's Inaugural Local AI Event Unveils Future Memory Roadmap
In a significant move indicative of its commitment to artificial intelligence (AI) and memory technology, Apple hosted its first-ever official local AI event at Apple Park on June 23rd. This private gathering featured notable executives including CEO Tim Cook and Senior Vice President of Hardware Engineering John Ternus, who addressed the future of on-device AI capabilities.
AI Advancements and Memory Configurations
Reported by Alex Cheema, the event unveiled several anticipated advancements in memory configurations, suggesting that Apple is not turning away from high-memory options despite the current challenges posed by DRAM shortages in the industry. According to Cheema, the company is likely to introduce two new chip variants: the M5 Ultra, which is expected to feature 512GB of unified memory, and the M7 Ultra, rumored to support up to 1.5TB of memory.
During the event, Cheema was unable to share the original slides but successfully recreated information that hints at Apple's ongoing investment in high-performance memory for various devices. The following table summarizes the memory configurations and the corresponding active parameters for Apple's devices as conveyed in the presentation:
| Device | Unified Memory (GB) | Active Parameters |
|---|---|---|
| iPhone and iPad | Up to 16 | Up to 14 billion |
| MacBook Air | Up to 32 | Up to 35 billion |
| Mac mini | Up to 64 | Up to 70 billion |
| MacBook Pro | Up to 128 | Up to 120 billion |
| Mac Studio | Up to 512 | Up to 480 billion |
| Mac Studio Cluster | Up to 2TB | 1.6 trillion+ |
Future Roadmap and AI Servers
In alignment with insights from Bloomberg's Mark Gurman, it appears that Apple is progressing towards the development of an AI server based on the M5 Ultra, codenamed J246. Additionally, a second-generation server chip based on the anticipated M7 Ultra is slated for release around 2029. Looking even farther ahead, plans for an M8 chip (codenamed "Soko") and a high-end Mac chip (codenamed "Cardinal") are projected to arrive by 2028, both utilizing a cutting-edge 1.4nm manufacturing process.
Conclusion
While the details shared during the local AI event remain unofficial, they signal Apple's robust dedication to enhancing local AI capabilities across its devices. As the tech giant navigates the complexities of memory technology amidst a global supply chain crisis, it is clear that Apple remains focused on providing innovative solutions to its user base, solidifying its position as a leader in the tech industry.
As developments unfold in the coming months and years, the landscape of AI and memory technology at Apple continues to evolve, promising a future rich with possibilities.
👨🏫 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
