Apple A20 Pro could deliver major AI performance gains without LPDDR6 RAM
Apple’s next-generation A20 Pro chip is shaping up to be a major upgrade for future iPhone models, especially in the area of on-device artificial intelligence. While some rival mobile processors may move toward LPDDR6 memory sooner, Apple appears ready to continue using LPDDR5X RAM for the A20 Pro. That might sound like a disadvantage at first, but new technical details suggest Apple has found other ways to significantly improve memory performance and AI efficiency.
The A20 Pro is expected to be Apple’s first chip built on a 2nm process, and it could power the iPhone 18 Pro and iPhone 18 Pro Max. Instead of relying on LPDDR6 memory, Apple may boost bandwidth by changing how memory is connected to the chip. The key upgrade is support for six-channel LPDDR5X memory with a 96-bit bus width, a notable jump from the older four-channel, 64-bit setup used in previous designs.
This matters because modern AI workloads are often limited not only by raw processing power but also by how quickly data can move between memory and the processor. On-device AI models need fast access to large amounts of data. If memory bandwidth is too limited, even a powerful chip can be forced to wait for data, reducing real-world performance. By widening the memory bus and increasing the number of channels, Apple can help the A20 Pro keep its processing units fed with data more efficiently.
The A20 Pro is also expected to feature a larger Neural Engine, which should further improve performance in AI-related tasks. This could benefit features such as image generation, advanced Siri capabilities, real-time language processing, photo editing, voice recognition, and other Apple Intelligence functions that run directly on the device. Keeping these tasks on the iPhone instead of relying heavily on cloud servers can improve privacy, reduce latency, and make AI tools feel more responsive.
Another important change is Apple’s reported move away from its older integrated packaging approach toward Wafer-Level Multi-Chip Module packaging, also known as WMCM. This packaging method is better suited for high-bandwidth designs and allows the memory and processor die to be arranged in a way that can handle larger data volumes more effectively.
Separating the DRAM from the A20 Pro die may also help with heat management. Better heat dissipation can lead to stronger sustained performance, especially during demanding tasks such as AI processing, gaming, video editing, and computational photography. Thermal efficiency is increasingly important for smartphones because chips are becoming more powerful while devices remain thin and compact.
Although the lack of LPDDR6 support may disappoint users expecting Apple to adopt the newest memory standard immediately, the A20 Pro’s rumored design shows that memory speed is not the only path to better performance. A wider 96-bit bus, six-channel LPDDR5X support, improved packaging, and a larger Neural Engine could combine to deliver a major leap in AI capabilities.
Apple has often adopted certain technologies later than competitors, but it typically focuses on optimizing the full hardware and software stack rather than chasing specifications alone. If these details prove accurate, the A20 Pro could become one of Apple’s most important mobile chip upgrades in years, particularly as the iPhone continues to shift toward more powerful on-device AI experiences.
