ollama create
# https://huggingface.co/second-state/Baichuan2-7B-Chat-GGUF/blob/main/Baichuan2-7B-Chat-Q5_K_M.gguf
wget https://huggingface.co/second-state/Baichuan2-7B-Chat-GGUF/resolve/main/Baichuan2-7B-Chat-Q5_K_M.gguf
cat >Modelfile<<EOF
FROM ./Baichuan2-7B-Chat-Q5_K_M.gguf
EOF
ollama create baichuan2-7b-chat -f Modelfile
# https://modelscope.cn/models/shaowenchen/baichuan2-13b-chat-gguf/summary
wget -O baichuan2-13b-chat.Q5_1.gguf "https://modelscope.cn/api/v1/models/shaowenchen/baichuan2-13b-chat-gguf/repo?Revision=master&FilePath=baichuan2-13b-chat.Q5_1.gguf"
cat >Modelfile<<EOF
FROM ./baichuan2-13b-chat.Q5_1.gguf
EOF
ollama create baichuan2-13b-chat -f Modelfile
# grok-1-GGUF
# https://huggingface.co/mradermacher/grok-1-GGUF
# https://huggingface.co/mradermacher/grok-1-i1-GGUF
wget https://huggingface.co/mradermacher/grok-1-GGUF/resolve/main/grok-1.Q4_K_S.gguf.part1of4
wget https://huggingface.co/mradermacher/grok-1-GGUF/resolve/main/grok-1.Q4_K_S.gguf.part2of4
wget https://huggingface.co/mradermacher/grok-1-GGUF/resolve/main/grok-1.Q4_K_S.gguf.part3of4
wget https://huggingface.co/mradermacher/grok-1-GGUF/resolve/main/grok-1.Q4_K_S.gguf.part4of4
## Git LFS 实时显示进度条方法
GIT_LFS_SKIP_SMUDGE=1 git clone https://huggingface.co/lmsys/vicuna-13b-v1.5
cd vicuna-13b-v1.5
git lfs pull
GIT_LFS_SKIP_SMUDGE=1 git clone https://huggingface.co/lmsys/vicuna-7b-v1.5
cd vicuna-7b-v1.5
git lfs pull
weighted/imatrix quants
VS static quants
ref
- 标签 llm