llama.cpp-cuda b10078-1
Port of Facebook's LLaMA model in C/C++ (with NVIDIA CUDA optimizations)
Jul 21
arch
6b7fd13d371be3b30f3bee8aca244c6b7a5ad92438efd43c6e4b2d01cf3880e7
O(P)H₄(Cm₃DbOs₂Po)Md(Pa)