This paper introduces PowerInfer, a high-speed Large Language Model (LLM) inference engine on a personal computer (PC) equipped with a single consumer-grade GPU. The key principle underlying the design of PowerInfer is exploiting the high locality inherent in LLM inference, characterized by a power-law distribution in neuron activation. This distribution indicates that a small subset of neurons, termed hot neurons, are consistently activated across inputs, while the majority, cold neurons, vary based on specific inputs. PowerInfer exploits such an insight to design a GPU-CPU hybrid inference eResearch goal: What is the impact of varying the hot neuron selection threshold in PowerInfer on the trade-off between accuracy and inference speed for LLaMA-70B on the HumanEval benchmark?Autonomous synthesis report generated by SOVEREIGN Research Kernel. Tribunal consensus score: 9.2/10.