Create Dockerfile

Dockerfile

@@ -0,0 +1,46 @@
+ARG UBUNTU_VERSION=22.04
+ARG CUDA_VERSION=12.3.1
+ARG BASE_CUDA_DEV_CONTAINER=nvidia/cuda:${CUDA_VERSION}-devel-ubuntu${UBUNTU_VERSION}
+ARG BASE_CUDA_RUN_CONTAINER=nvidia/cuda:${CUDA_VERSION}-runtime-ubuntu${UBUNTU_VERSION}
+
+FROM ${BASE_CUDA_DEV_CONTAINER} as build
+
+ARG CUDA_DOCKER_ARCH=all
+
+RUN apt-get update --fix-missing && \
+    apt-get install -y --no-install-recommends git build-essential gcc cmake && \
+    rm -rf /var/lib/apt/lists/*
+
+WORKDIR /build
+
+RUN git clone https://github.com/ggerganov/llama.cpp.git
+
+WORKDIR /build/llama.cpp
+
+ENV CUDA_DOCKER_ARCH=${CUDA_DOCKER_ARCH}
+ENV LLAMA_CUBLAS=1
+
+RUN mkdir build && \
+    cd build && \
+    cmake .. -DLLAMA_CUBLAS=ON && \
+    cmake --build . --config Release
+
+FROM ${BASE_CUDA_RUN_CONTAINER} as runtime
+RUN apt-get update --fix-missing && \
+    apt-get install -y --no-install-recommends wget && \
+    rm -rf /var/lib/apt/lists/*
+
+WORKDIR /app
+
+# Copy the executable from the build stage
+COPY --from=build /build/llama.cpp/build/bin/server /app
+COPY --from=build /build/llama.cpp/examples/server/public /app/public
+COPY ./run.sh /app/run.sh
+WORKDIR /app
+EXPOSE 7867
+
+# Make the script executable
+RUN chmod +x run.sh
+
+# CMD to run your script
+CMD ./run.sh