Apple has officially announced the M5 chip, representing the next major leap in Apple Silicon performance and efficiency. Built on third-generation 3-nanometer technology, M5 introduces a next-generation 10-core GPU with a Neural Accelerator in every core, delivering more than four times the peak GPU compute performance for AI compared to M4, and over six times faster than M1. The chip also features a redesigned CPU with up to 10 cores – including four performance cores and six efficiency cores – that together enable the world’s fastest performance core for both everyday computing and AI-intensive tasks.
According to Apple, the M5 chip powers the new 14-inch MacBook Pro, iPad Pro, and Apple Vision Pro, all of which are available for pre-order starting today. The CPU delivers up to 15% faster multithreaded performance than M4, while the GPU achieves up to 30% higher overall graphics performance and as much as 45% faster ray-tracing. Unified memory bandwidth has also been increased by nearly 30%, reaching 153GB/s, and the chip supports configurations with up to 32GB of unified memory for running demanding AI and creative workloads locally on device.
Johny Srouji, Apple’s senior vice president of Hardware Technologies, said that “M5 ushers in the next big leap in AI performance for Apple Silicon,” highlighting how Neural Accelerators embedded in each GPU core dramatically accelerate machine learning operations. Developers using Apple’s frameworks – including Core ML, Metal Performance Shaders, and Metal 4 – can benefit immediately from the improved performance. For more advanced use cases, Metal 4’s Tensor APIs allow direct programming of the Neural Accelerators to optimize AI workflows.
The new GPU architecture was built specifically for both AI and graphics performance. With enhanced shader cores, third-generation ray tracing, and rearchitected dynamic caching, the M5 enables smoother gameplay, faster rendering, and more realistic visuals in creative and 3D applications. On Apple Vision Pro, this results in 10% more rendered pixels on the micro-OLED displays and up to 120Hz refresh rates for crisper visuals, reduced motion blur, and a more fluid experience. Apple says this performance uplift benefits apps that rely on AI-driven rendering or large-scale simulation workloads.
In real-world examples, the M5 accelerates AI workflows such as running diffusion models in Draw Things or executing large language models locally using webAI. This architecture allows developers and users to take advantage of on-device AI without relying on cloud infrastructure, preserving privacy while boosting responsiveness. For professionals, the combination of faster GPU compute, high bandwidth memory, and efficient CPU design allows smoother multitasking between heavy creative applications like Final Cut Pro, Adobe Photoshop, and large file uploads.
The 16-core Neural Engine in M5 is also faster and more efficient, complementing the Neural Accelerators in the CPU and GPU. It enhances Apple Intelligence features such as Image Playground and Genmoji, and speeds up AI functions on Apple Vision Pro like transforming 2D photos into spatial scenes or generating Personas. Developers using Apple’s Foundation Models framework can also expect improved training and inference speeds due to higher unified memory bandwidth and faster on-device model execution.
Apple emphasized that the M5’s efficiency plays a critical role in supporting its broader environmental goals. As part of the Apple 2030 initiative, which aims to achieve carbon neutrality across the company’s entire footprint by the end of this decade, the M5 enables new levels of energy efficiency across MacBook Pro, iPad Pro, and Apple Vision Pro. By reducing power consumption while increasing performance, the chip contributes to Apple’s plan to lower the lifetime energy impact of its products.
With its advanced AI-optimized GPU, world’s fastest performance core, faster Neural Engine, and improved unified memory system, the M5 chip represents Apple’s most significant upgrade yet for AI and graphics performance. It redefines what users and developers can do on Apple devices, combining power efficiency, on-device intelligence, and professional-grade computing in a single, unified architecture.
2 comments