Update Dockerfile

Dockerfile

@@ -26,7 +26,7 @@ ENV PATH="/usr/local/cuda/bin:$PATH" \
 WORKDIR /app
 
 # Download ggml and mmproj models from HuggingFace
-RUN wget https://huggingface.co/mys/ggml_llava-v1.5-13b/resolve/main/ggml-model-q4_k.gguf && \
+RUN wget https://huggingface.co/mys/ggml_llava-v1.5-13b/resolve/main/ggml-model-q5_k.gguf && \
     wget https://huggingface.co/mys/ggml_llava-v1.5-13b/resolve/main/mmproj-model-f16.gguf 
 
 # Clone and build llava-server with CUDA support
@@ -42,7 +42,7 @@ RUN useradd -m -u 1000 user && \
     cp /app/ggml-model-q4_k.gguf /home/user/app && \
     cp /app/mmproj-model-f16.gguf /home/user/app
 
-RUN chown user:user /home/user/app/ggml-model-q4_k.gguf && \
+RUN chown user:user /home/user/app/ggml-model-q5_k.gguf && \
     chown user:user /home/user/app/mmproj-model-f16.gguf
 
 USER user
@@ -54,4 +54,4 @@ WORKDIR $HOME/app
 EXPOSE 8080
 
 # Start the llava-server with models
-CMD ["/app/llava-cpp-server/bin/llava-server", "-m", "ggml-model-q4_k.gguf", "--mmproj", "mmproj-model-f16.gguf", "--host", "0.0.0.0"]
+CMD ["/app/llava-cpp-server/bin/llava-server", "-m", "ggml-model-q5_k.gguf", "--mmproj", "mmproj-model-f16.gguf", "--host", "0.0.0.0"]