Summary: PyTorch now optimized for Apple’s new chip
Apple’s transition to its own processors, called “Apple Silicon,” has been bumpy, particularly for users working with large language models (LLMs), due to the need for software developers to adjust programs to make full use of the new architecture. Support for neural networks on Apple’s new GPUs remained a sticking point for users. However, PyTorch recently released version 2.0 which ships with more Metal Performance Shader (MPS) support, significantly improving the ability for LLMs to run on the new Apple Silicon products. Running LLMs on consumer laptops is now a possibility, putting the full capabilities of the new chip within reach.
Related articles
Large Language Model Inference with PyTorch on Apple Silicon
More than two years ago, Apple began its transition away from Intel processors to their own chips: Apple Silicon. The transition has been a sometimes …
Read the complete article at: www.hendrik-erz.de
About The Author
