Fix

Dockerfile

@@ -21,10 +21,11 @@ WORKDIR /app
 RUN apt-get update && \
     apt-get install -y --no-install-recommends \
       build-essential cmake git curl wget ninja-build libgomp1 ca-certificates \
-      gcc g++ libffi-dev libopenblas-dev libstdc++6 libgcc-s1 && \
-    rm -rf /var/lib/apt/lists/* && \
-    curl --proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs | sh -s -- -y && \
-    rustup default stable
+      gcc g++ libffi-dev libopenblas-dev libstdc++6 libgcc-s1 \
+      libcurl4-openssl-dev \                  # ← add this
+    && rm -rf /var/lib/apt/lists/* \
+    && curl --proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs | sh -s -- -y \
+    && rustup default stable
 
 # 5) Prepare directories
 RUN mkdir -p /app/cache /app/pretrained_models/llm && \
@@ -45,7 +46,7 @@ RUN git clone --recursive https://github.com/abetlen/llama-cpp-python.git /tmp/l
 # 8) Install huggingface_hub CLI
 RUN pip install --no-cache-dir huggingface_hub
 
-# 9) Clone llama.cpp for quantize tool
+# 9) Clone llama.cpp and build just the quantize tool
 RUN git clone https://github.com/ggerganov/llama.cpp.git /tmp/llama.cpp && \
     cd /tmp/llama.cpp && \
     cmake -B build -DLLAMA_BUILD_QUANTIZE=ON . && \
@@ -55,11 +56,14 @@ RUN git clone https://github.com/ggerganov/llama.cpp.git /tmp/llama.cpp && \
 RUN python3 - <<EOF
 import os
 from huggingface_hub import login, hf_hub_download
-if os.getenv('HF_TOKEN'): login(token=os.getenv('HF_TOKEN'))
-hf_hub_download(repo_id=os.getenv('MODEL_REPO'),
-                filename=os.getenv('MODEL_FILE'),
-                local_dir='/app/pretrained_models/llm',
-                token=os.getenv('HF_TOKEN'))
+if os.getenv('HF_TOKEN'):
+    login(token=os.getenv('HF_TOKEN'))
+hf_hub_download(
+    repo_id=os.getenv('MODEL_REPO'),
+    filename=os.getenv('MODEL_FILE'),
+    local_dir='/app/pretrained_models/llm',
+    token=os.getenv('HF_TOKEN')
+)
 EOF
 
 RUN /tmp/llama.cpp/build/quantize \