RedToasty/llama.cpp_qts: Fixing --split-mode tensor, with different KV cache quantization types. (opens in new tab)
Fixing --split-mode tensor, with different KV cache quantization types. - RedToasty/llama.cpp_qts
Read the original articleFixing --split-mode tensor, with different KV cache quantization types. - RedToasty/llama.cpp_qts
Read the original article