Refactor start.sh: streamline model download and argument building

Co-authored-by: Copilot <copilot@github.com>

tooling/pipeline/runpod/Dockerfile

@@ -1,65 +1,72 @@
-# Phase 1: Pull prebuilt binaries
-FROM ghcr.io/ggml-org/llama.cpp:full-cuda AS llama-bin
-
-# Phase 2: Building environment
-FROM nvidia/cuda:12.6.3-devel-ubuntu24.04
+# --- Stage 1: Build Environment (The "Heavy" Stage) ---
+FROM nvidia/cuda:12.6.3-devel-ubuntu24.04 AS builder
 
 ENV DEBIAN_FRONTEND=noninteractive \
-  CMAKE_GENERATOR=Ninja \
-  APP_ROOT=/workspace/app \
-  BUILD_DIR=/workspace/app/build
+  CMAKE_GENERATOR=Ninja
 
 RUN apt-get update && apt-get install -y --no-install-recommends \
-  build-essential \
-  ca-certificates \
-  curl \
-  git \
-  libboost-json-dev \
-  libboost-program-options-dev \
-  libssl-dev \
-  ninja-build \
-  pkg-config \
-  zlib1g-dev \
+  build-essential ca-certificates curl git libboost-json-dev \
+  libboost-program-options-dev libssl-dev ninja-build pkg-config zlib1g-dev \
   && rm -rf /var/lib/apt/lists/*
 
-# Install modern CMake via curl (Ubuntu 24.04 'apt' version can be laggy)
+# Install modern CMake
 RUN curl -L https://github.com/Kitware/CMake/releases/download/v3.31.0/cmake-3.31.0-linux-x86_64.sh -o cmake.sh && \
   sh cmake.sh --skip-license --prefix=/usr/local && rm cmake.sh
 
-# Copy backends to /usr/local/lib and register with ldconfig so the
-# runtime linker can resolve libllama.so, libggml.so, libggml-base.so etc.
-COPY --from=llama-bin /app/lib*.so* /usr/local/lib/
-RUN ldconfig
+# Get headers for C++ build
+RUN curl -L https://github.com/ggml-org/llama.cpp/archive/refs/tags/b9012.tar.gz -o /tmp/llama-src.tar.gz && \
+  tar -xzf /tmp/llama-src.tar.gz -C /tmp && \
+  cp -r /tmp/llama.cpp-b9012/include/* /usr/local/include/ && \
+  cp -r /tmp/llama.cpp-b9012/ggml/include/* /usr/local/include/
 
-# Headers for C++ Build
-RUN git clone --depth 1 -b b9012 https://github.com/ggml-org/llama.cpp.git /tmp/llama-src && \
-  cp -r /tmp/llama-src/include/* /usr/local/include/ && \
-  cp -r /tmp/llama-src/ggml/include/* /usr/local/include/ && \
-  rm -rf /tmp/llama-src
+# Pull llama.cpp binaries to use during build if needed
+COPY --from=ghcr.io/ggml-org/llama.cpp:full-cuda /app/lib*.so* /usr/local/lib/
 
-ENV LD_LIBRARY_PATH="/usr/local/lib:${LD_LIBRARY_PATH}"
-
-WORKDIR /workspace/app
+WORKDIR /app
 COPY . .
 
 # Build the C++ pipeline
 RUN cmake -S . -B build -G Ninja -DCMAKE_BUILD_TYPE=Release && \
   cmake --build build -j$(nproc)
 
-# Co-locate GGML backend plugins with the executable.
-# ggml_backend_load_all() searches the executable directory first when
-# GGML_BACKEND_DIR is not set. Copying the ggml-*.so plugin files here
-# ensures the loader finds them without any environment variable.
-# libllama.so, libggml.so, and libggml-base.so are NOT copied here —
-# those are proper shared libraries resolved via ldconfig/LD_LIBRARY_PATH.
-RUN cp /usr/local/lib/libggml-cuda.so  /workspace/app/build/ 2>/dev/null || true && \
-  cp /usr/local/lib/libggml-cpu*.so   /workspace/app/build/ 2>/dev/null || true && \
-  cp /usr/local/lib/libggml-blas*.so  /workspace/app/build/ 2>/dev/null || true && \
-  cp /usr/local/lib/libggml-rpc*.so   /workspace/app/build/ 2>/dev/null || true
+# --- Stage 2: Runtime Environment (The "Slim" Stage) ---
+FROM nvidia/cuda:12.6.3-runtime-ubuntu24.04 AS runtime
+
+# Install only necessary runtime shared libraries
+RUN apt-get update && apt-get install -y --no-install-recommends \
+  curl \
+  ca-certificates \
+  libboost-json1.83.0 \
+  libboost-program-options1.83.0 \
+  libgomp1 \
+  libssl3 \
+  zlib1g \
+  && rm -rf /var/lib/apt/lists/*
+
+ENV APP_ROOT=/app \
+  LD_LIBRARY_PATH="/usr/local/lib:${LD_LIBRARY_PATH}"
+
+WORKDIR /app/build
+
+# Copy only the compiled binaries from the builder
+COPY --from=builder /app/build/biergarten-pipeline ./
+
+# Copy required config files
+COPY locations.json /app/build/
+COPY beer-styles.json /app/build/
+
+# Copy prompt templates
+COPY prompts /app/prompts
+
+# Copy only the necessary shared libraries from builder/llama-bin
+COPY --from=ghcr.io/ggml-org/llama.cpp:full-cuda /app/lib*.so* /usr/local/lib/
+
+# Co-locate plugins
+RUN cp /usr/local/lib/libggml-cuda.so . 2>/dev/null || true && \
+  cp /usr/local/lib/libggml-cpu*.so . 2>/dev/null || true
 
 # Setup Start Script
-COPY runpod/start.sh /usr/local/bin/biergarten-start
+COPY ./runpod/start.sh /usr/local/bin/biergarten-start
 RUN chmod +x /usr/local/bin/biergarten-start
 
-WORKDIR /workspace/app/build
 ENTRYPOINT ["/usr/local/bin/biergarten-start"]

tooling/pipeline/runpod/start.sh

@@ -1,49 +1,58 @@
 #!/bin/bash
 set -e
 
-# Configuration / Defaults
 MODEL_PATH="${BIERGARTEN_MODEL_PATH:-/workspace/models/google_gemma-4-E4B-it-Q6_K.gguf}"
 OUTPUT_DIR="${BIERGARTEN_OUTPUT_DIR:-/workspace/output}"
 LOG_PATH="${BIERGARTEN_LOG_PATH:-/workspace/logs/pipeline.log}"
-EXECUTABLE="/workspace/app/build/biergarten-pipeline"
-PROMPT_DIR="/workspace/app/build/prompts"
+EXECUTABLE="/app/build/biergarten-pipeline"
+PROMPT_DIR="/app/prompts"
 
 echo "--- Starting Biergarten Pipeline Environment Check ---"
 
-# 1. Ensure volume mount directories exist
+# Ensure directories exist
 mkdir -p "$OUTPUT_DIR"
 mkdir -p "$(dirname "$LOG_PATH")"
+mkdir -p "$(dirname "$MODEL_PATH")"
 
-# 2. Check for model file
+# Download model if missing
 if [ ! -f "$MODEL_PATH" ]; then
-    echo "ERROR: Model not found at $MODEL_PATH"
-    echo "Current /workspace/models contents:"
-    ls -lh /workspace/models 2>/dev/null || echo "(directory does not exist)"
+    echo "Model not found. Downloading (this may take a while)..."
+
+    curl -L -C - \
+      -o "$MODEL_PATH" \
+      "https://huggingface.co/bartowski/google_gemma-4-E4B-it-GGUF/resolve/main/google_gemma-4-E4B-it-Q6_K.gguf?download=true"
+
+    echo "Download complete."
+fi
+
+# Verify model exists
+if [ ! -f "$MODEL_PATH" ]; then
+    echo "ERROR: Model still not found after download attempt."
     exit 1
 fi
 
-# 3. Build the command arguments
+# Default GPU layers
+GL_LAYERS="${BIERGARTEN_GL_LAYERS:-40}"
+
+# Build args
 ARGS=(
     "--model"      "$MODEL_PATH"
     "--prompt-dir" "$PROMPT_DIR"
     "--output"     "$OUTPUT_DIR"
     "--log-path"   "$LOG_PATH"
+    "--n-gpu-layers" "$GL_LAYERS"
 )
 
-# Optional hyperparameters
+# Optional params
 [[ -n "$BIERGARTEN_TEMPERATURE" ]] && ARGS+=("--temperature"  "$BIERGARTEN_TEMPERATURE")
 [[ -n "$BIERGARTEN_TOP_P" ]]       && ARGS+=("--top-p"        "$BIERGARTEN_TOP_P")
 [[ -n "$BIERGARTEN_TOP_K" ]]       && ARGS+=("--top-k"        "$BIERGARTEN_TOP_K")
 [[ -n "$BIERGARTEN_N_CTX" ]]       && ARGS+=("--n-ctx"        "$BIERGARTEN_N_CTX")
 [[ -n "$BIERGARTEN_SEED" ]]        && ARGS+=("--seed"         "$BIERGARTEN_SEED")
-[[ -n "$BIERGARTEN_GL_LAYERS" ]]   && ARGS+=("--n-gpu-layers" "$BIERGARTEN_GL_LAYERS")
 
-# Append any extra custom args
-if [[ -n "$BIERGARTEN_EXTRA_ARGS" ]]; then
-    ARGS+=($BIERGARTEN_EXTRA_ARGS)
-fi
+# Extra args
+[[ -n "$BIERGARTEN_EXTRA_ARGS" ]] && ARGS+=($BIERGARTEN_EXTRA_ARGS)
 
 echo "--- Executing: $EXECUTABLE ${ARGS[*]} ---"
 
-# Execute the binary directly, replacing the shell process
 exec "$EXECUTABLE" "${ARGS[@]}"