Update Dockerfile

Dockerfile

@@ -1,63 +1,18 @@
-# Use Ubuntu as base image
-FROM ubuntu:22.04
-
-# Set working directory
-WORKDIR /workspaces/empty
-
-# Update package list and install required dependencies
-RUN apt-get update && apt-get install -y \
-    wget \
-    unzip \
-    curl \
-    libgomp1 \
-    libopenblas0 \
-    libc6 \
-    libgcc-s1 \
-    libstdc++6 \
-    && rm -rf /var/lib/apt/lists/*
-
-# Step 1: Download llama.cpp Precompiled Binary
-RUN wget https://github.com/ggml-org/llama.cpp/releases/download/b5452/llama-b5452-bin-ubuntu-x64.zip
-
-# Step 2: Unzip and examine contents
-RUN unzip llama-b5452-bin-ubuntu-x64.zip && ls -la
-
-# Step 3: Create build directory and move binaries
-RUN mkdir -p /workspaces/empty/build/bin && \
-    if [ -d "llama-b5452-bin-ubuntu-x64" ]; then \
-        mv llama-b5452-bin-ubuntu-x64/* /workspaces/empty/build/bin/; \
-        rmdir llama-b5452-bin-ubuntu-x64; \
-    else \
-        find . -maxdepth 1 -name "*llama*" -type f -executable -exec mv {} /workspaces/empty/build/bin/ \; ; \
-    fi
-
-# Make sure binaries are executable
-RUN chmod +x /workspaces/empty/build/bin/*
-
-# Create cache directory and set permissions
-RUN mkdir -p /.cache/llama.cpp && \
-    chmod -R 777 /.cache
-
-# Create a non-root user for running the application
-RUN groupadd -r llama && useradd -r -g llama llama && \
-    mkdir -p /home/llama/.cache/llama.cpp && \
-    chown -R llama:llama /home/llama
-
-# Set environment variable for cache directory
-ENV XDG_CACHE_HOME=/home/llama/.cache
-
-# Switch to non-root user
-USER llama
-
-# Step 4 & 5: Set Environment Variables for PATH and LD_LIBRARY_PATH
-ENV PATH="/workspaces/empty/build/bin:${PATH}"
-ENV LD_LIBRARY_PATH="/workspaces/empty/build/bin:${LD_LIBRARY_PATH}"
-
-# Clean up downloaded zip file
-RUN rm -f llama-b5452-bin-ubuntu-x64.zip
-
-# Expose the default port for llama-server (usually 8080)
-EXPOSE 8080
-
-# Step 6: Set the default command to run llama-server
-CMD ["llama-server", "-hf", "ggml-org/SmolVLM-500M-Instruct-GGUF"]
+    FROM ubuntu:latest
+
+    RUN apt-get update && apt-get install -y git build-essential wget
+
+    WORKDIR /app
+
+    RUN git clone https://github.com/ggerganov/llama.cpp.git
+
+    WORKDIR /app/llama.cpp
+
+    RUN make
+
+    # Download model - replace with your model URL or local path
+    RUN wget -O models/7B/ggml-model-q4_0.bin https://huggingface.co/TheBloke/Llama-2-7B-GGML/resolve/main/llama-2-7b.ggmlv3.q4_0.bin
+
+    EXPOSE 8080
+
+    CMD ["./server", "--model", "models/7B/ggml-model-q4_0.bin", "--port", "8080", "--host", "0.0.0.0"]