We read every piece of feedback, and take your input very seriously.
To see all available qualifiers, see our documentation.
context
load_duration
total_duration
GGML_CUDA_FORCE_MMQ=on