Comment on Apple Silicon and macOS VMs: Faster LLM Inference with llama.cppparentComments−frabonacciOP1momy guess is apple chose a conservative profile for compatibility across different chips and guest releases
Comments
my guess is apple chose a conservative profile for compatibility across different chips and guest releases