Apple unveils M6 and M5 Ultra chips with major AI performance gains
Apple says M6 delivers nearly 30% higher peak GPU AI compute than M5, while M5 Ultra provides up to 4.5x the peak GPU AI compute of M3 Ultra.
Apple has introduced its latest desktop silicon with the M6 and M5 Ultra, emphasizing on-device AI alongside gains in CPU, GPU and memory performance.
The M6 is Apple’s first 2-nanometer chip and adds two CPU and GPU cores versus M5, a Dual 16-core Neural Engine and up to 170GB/s of unified memory bandwidth. M5 Ultra combines four dies through next-generation UltraFusion technology, reaching up to 36 CPU cores, 80 GPU cores, 512GB of unified memory and 1.2TB/s of bandwidth.
The new architecture is geared toward increasingly intensive AI workloads. M6 offers nearly 30% more peak GPU AI compute than M5 and supports up to 32GB of unified memory for running LLMs and agentic AI tasks locally.
M5 Ultra combines its larger GPU, Neural Accelerators and massive memory pool to deliver up to 4.5x the peak GPU AI compute of M3 Ultra and support local execution of models with hundreds of billions of parameters.
AI, tech, and the markets they move—in one daily briefing.
Daily. Free. Join 34,000+ readers across crypto, finance, and policy.
The new chips also target professional workloads including video editing, 3D rendering, scientific analysis, coding and agentic AI.
Apple says M5 Ultra can run huge LLMs with hundreds of billions of parameters entirely on-device, while M6 is designed to accelerate everyday AI tasks and local model execution. Developers can access the hardware through Apple’s Core AI, Core ML, Metal and Xcode frameworks.