update to latest 0.1.52 llama.cpp

.github/workflows/build-llama-cpp-wheel.yml

@@ -26,8 +26,8 @@ jobs:
 
       - name: Copy GPU & CPU artifact from Docker container
         run: |
-          docker cp my-artifact-builder:/build/dists/llama_cpp_python-gpu-0.1.50-cp38-cp38-linux_x86_64.whl ./llama_cpp_python-gpu-0.1.50-cp38-cp38-linux_x86_64.whl
-          docker cp my-artifact-builder:/build/dists/llama_cpp_python-cpu-0.1.50-cp38-cp38-linux_x86_64.whl ./llama_cpp_python-cpu-0.1.50-cp38-cp38-linux_x86_64.whl
+          docker cp my-artifact-builder:/build/dists/llama_cpp_python-gpu-0.1.52-cp38-cp38-linux_x86_64.whl ./llama_cpp_python-gpu-0.1.52-cp38-cp38-linux_x86_64.whl
+          docker cp my-artifact-builder:/build/dists/llama_cpp_python-cpu-0.1.52-cp38-cp38-linux_x86_64.whl ./llama_cpp_python-cpu-0.1.52-cp38-cp38-linux_x86_64.whl
 
       - name: Upload artifacts
         uses: actions/upload-artifact@v3
@@ -56,4 +56,4 @@ jobs:
         with:
           files: |
             *.whl
-          token: ${{ secrets.GITHUB_TOKEN }}
+          token: ${{ secrets.GITHUB_TOKEN }}

Dockerfile-llama-cpp-wheel

@@ -1,6 +1,6 @@
 FROM nvidia/cuda:11.3.1-cudnn8-devel-ubuntu18.04
 
-ARG LLAMA_CPP_VERSION="0.1.50"
+ARG LLAMA_CPP_VERSION="0.1.52"
 ARG CMAKE_VERSION=3.26
 ARG CMAKE_VERSION_PATCH=3.26.3
 ARG CMAKE_OS=linux