mirror of
https://github.com/ggml-org/llama.cpp.git
synced 2026-09-18 00:34:55 +02:00
86df2c9ae4
* use a more flexible amount of threads * fix windows compile and 0 thread case * nominmax