Closed RakshitAralimatti closed 3 months ago
The tool has been renamed from quantize
to llama-quantize
Thanks for your response. Now its working
@ggerganov llama-quantize: failed to access
showed the message as follows; /bin/bash: line 1: ./llama.cpp/llama-quantize: No such file or directory
Run make
first
@ggerganov Run make as follows:
!git clone https://github.com/ggerganov/llama.cpp !cd llama.cpp && git pull && make clean && LLAMA_CUBLAS=1 make !pip install -r llama.cpp/requirements.txt
Is it coming correctly?
What happened?
For converting the FP16.gguf to q5_k_m for llama3 i was getting the error sh: 1: ./llama.cpp/llama-quantize: not found previously i used os.system("./llama.cpp/llama-quantize " + gguf_dir + "/" + gguf_F16_name + " " + model_path + " " + m) in my script to convert even that is not working now its giving the same not found issue
Name and Version
Lastest pervsion
What operating system are you seeing the problem on?
Linux
Relevant log output