llama.cpp
Leverage the existing GGML_F32_VEC helpers to vectorize ggml_vec_set_f32 for faster fills
#16522
Merged
Merged