Hardware
What is GPU Cores?
The parallel processing units in Apple Silicon used for LLM inference via the Metal framework. More GPU cores generally means faster tok/s. The M5 Max has 40 GPU cores, the M4 Pro has 20, and the base M3 has 10. According to the LLMCheck index, GPU core count correlates with inference speed but memory bandwidth is the stronger predictor.
Where GPU Cores comes up on LLMCheck
- Getting Started with MLX: Apple's AI Framework for Mac
- M4 Max vs M3 Max for Local LLM: Is the Upgrade Worth It?
- M5 Max MacBook Pro vs. M4 Max Mac Studio: The Local LLM Showdown
- M5 Pro vs M5 Max for Local LLM: Which MacBook Pro to Buy?
- Apple Silicon Neural Engine Explained: How Your Mac Runs AI
- Apple Silicon Memory Bandwidth & LLM Speed (2026): M1–M6, Pro, Max, Ultra
See the best local LLMs for every Mac →