Apple today announced its latest silicon breakthrough, the Apple M5, a chip designed to…
Apple today announced its latest silicon breakthrough, the Apple M5, a chip designed to usher in a new generation of on-device AI performance and graphics power. Built on third-generation 3nm technology, the M5 introduces a next-generation GPU architecture with Neural Accelerators embedded in every core, along with significant upgrades to the CPU, Neural Engine, and memory system, delivering over four times the peak GPU compute for AI compared to M4.
At the heart of M5 is a 10-core GPU, where each core integrates a dedicated Neural Accelerator. This architecture allows AI workloads such as inference, deep learning, and image processing to run directly on the GPU with remarkable efficiency. Apple claims M5 achieves over 30 to 45 percent higher graphics performance than M4 when handling shader workloads and ray tracing, while supporting third-generation ray tracing for richer, more realistic visuals.
The CPU also sees a major leap forward with up to ten cores — six efficiency cores and four performance cores — providing up to 15 percent faster multithreaded performance compared to M4. Complementing this is a faster 16-core Neural Engine designed for AI operations, which enhances both performance and energy efficiency across the system.
Another key advancement is the increase in unified memory bandwidth to 153GB/s, nearly 30 percent more than M4. This enables the CPU, GPU, and Neural Engine to access a shared pool of memory simultaneously and at higher speeds, allowing users to run larger AI models, handle heavier creative workloads, and multitask between resource-intensive apps without lag.
