File size: 310 Bytes
3eaaf3a
 
 
 
 
 
 
 
 
 
 
 
e510c4e
3eaaf3a
e88cde7
e510c4e
3eaaf3a
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
#!/bin/bash

if [ ! -d "llama.cpp" ]; then
  # only run in dev env
  git clone https://github.com/ggerganov/llama.cpp
fi

if [[ -z "${RUN_LOCALLY}" ]]; then
  # enable CUDA if NOT running locally
  export LLAMA_CUDA=1
fi

cd llama.cpp
make -j llama-quantize llama-gguf-split llama-imatrix

cd ..
python app.py