Update Dockerfile

Dockerfile

@@ -1,6 +1,7 @@
 FROM nvidia/cuda:12.2.0-devel-ubuntu20.04
 
 ENV DEBIAN_FRONTEND=noninteractive
+
 # Install dependencies
 RUN apt update && \
     apt install --no-install-recommends -y build-essential python3 python3-pip wget curl git cmake zlib1g-dev && \
@@ -9,8 +10,8 @@ RUN apt update && \
 WORKDIR /app
 
 # Download ggml and mmproj models from HuggingFace
-RUN wget https://huggingface.co/mys/ggml_llava-v1.5-13b/raw/main/ggml-model-q4_k.gguf -O ggml-model-q4_k.gguf && \
-    wget https://huggingface.co/mys/ggml_llava-v1.5-13b/raw/main/mmproj-model-f16.gguf -O mmproj-model-f16.gguf
+RUN wget https://huggingface.co/mys/ggml_llava-v1.5-13b/raw/main/ggml-model-q4_k.gguf && \
+    wget https://huggingface.co/mys/ggml_llava-v1.5-13b/raw/main/mmproj-model-f16.gguf 
 
 # Clone and build llava-server
 RUN git clone https://github.com/matthoffner/llava-cpp-server.git llava && \
@@ -20,7 +21,10 @@ RUN git clone https://github.com/matthoffner/llava-cpp-server.git llava && \
     make
 
 # Create a non-root user for security reasons
-RUN useradd -m -u 1000 user
+RUN useradd -m -u 1000 user && \
+    mkdir -p /home/user/app && \
+    cp /app/ggml-model-q4_k.gguf /home/user/app && \
+    cp /app/mmproj-model-f16.gguf /home/user/app
 
 USER user
 ENV HOME=/home/user
@@ -31,4 +35,4 @@ WORKDIR $HOME/app
 EXPOSE 8080
 
 # Start the llava-server with models
-CMD ["./llava/bin/llava-server", "-m", "./ggml-model-q4_k.gguf", "--mmproj", "./mmproj-model-f16.gguf", "--host", "0.0.0.0", "--port", "8080"]
+CMD ["../llava/bin/llava-server", "-m", "ggml-model-q4_k.gguf", "--mmproj", "mmproj-model-f16.gguf", "--host", "0.0.0.0", "--port", "8080"]