llama.cpp

e34d6d03 - convert_hf_to_gguf.py: refactor modify_tensors to call super (#18866)