Update Dockerfile

Dockerfile

@@ -30,8 +30,8 @@ RUN wget https://huggingface.co/mys/ggml_bakllava-1/resolve/main/ggml-model-q4_k
     wget https://huggingface.co/mys/ggml_bakllava-1/resolve/main/mmproj-model-f16.gguf 
 
 # Clone and build llava-server with CUDA support
-RUN git clone https://github.com/matthoffner/llava-cpp-server.git && \
-    cd llava-cpp-server && \
+RUN git clone https://github.com/ggerganov/llama.cpp.git && \
+    cd llama.cpp && \
     git submodule init && \
     git submodule update && \
     make LLAMA_CUBLAS=1
@@ -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", "--threads", "10"]
+CMD ["/app/llama-cpp/server", "-m", "ggml-model-q4_k.gguf", "--mmproj", "mmproj-model-f16.gguf", "--host", "0.0.0.0", "--threads", "10"]