Upload bloomz-7b1-q4_0.gguf, converted from bloomz-7b1 by llama.cpp's bash convert-hf-to-gguf.py.
python convert-hf-to-gguf.py --outfile bloomz-7b1.gguf --outtype f16 /mnt/disk1/models/bloomz-7b1/
./build/bin/quantize bloomz-7b1.gguf bloomz-7b1-q4_0.gguf q4_0