fix: optimize backend Dockerfile with multi-stage build and pre-built wheels

Dockerfile

@@ -1,35 +1,50 @@
-# Use the official Python 3.10 slim image
-FROM python:3.10-slim
-
-# Set the working directory to /app
+# Stage 1: Builder
+# We use a standard Python image and install uv, which is safer and still incredibly fast.
+FROM python:3.10-slim-bookworm AS builder
 WORKDIR /app
 
-# Copy the requirements file into the container at /app
+# Install uv (The extremely fast Python package installer)
+RUN pip install uv
+
+# Configure uv
+ENV UV_COMPILE_BYTECODE=1
+ENV UV_LINK_MODE=copy
+
+# Copy requirements
 COPY requirements.txt .
 
-# Install system dependencies required for building llama-cpp-python
-RUN apt-get update && apt-get install -y \
-    build-essential \
-    cmake \
-    && rm -rf /var/lib/apt/lists/*
+# Create a virtual environment and install dependencies using uv
+# We point to the extra index URL for the pre-built llama-cpp-python wheels
+RUN uv venv /app/.venv && \
+    uv pip install \
+    --no-cache \
+    -r requirements.txt \
+    --extra-index-url https://abetlen.github.io/llama-cpp-python/whl/cpu \
+    --python /app/.venv
+
+# Stage 2: Final Runtime Image
+FROM python:3.10-slim-bookworm
+WORKDIR /app
+
+# Copy the virtual environment from the builder stage
+COPY --from=builder /app/.venv /app/.venv
 
-# Install dependencies
-RUN pip install --no-cache-dir -r requirements.txt
+# Set environment variables to use the virtual environment automatically
+ENV PATH="/app/.venv/bin:$PATH"
 
-# Copy the current directory contents into the container at /app
+# Copy the application code
 COPY . .
 
-# Create a non-root user and switch to it (required by Hugging Face Spaces)
+# Create a non-root user for security (Hugging Face Spaces requirement)
 RUN useradd -m -u 1000 user
 USER user
-ENV HOME=/home/user \
-    PATH=/home/user/.local/bin:$PATH
+ENV HOME=/home/user
 
-# Expose port 7860 (Hugging Face Spaces default)
+# Expose the application port
 EXPOSE 7860
 
-# Set context size for Hugging Face Spaces (Pure 16GB RAM)
+# Runtime configuration
 ENV MODEL_CTX_SIZE=8192
 
-# Run uvicorn when the container launches
-CMD ["uvicorn", "main:app", "--host", "0.0.0.0", "--port", "7860"]
+# Run the application
+CMD ["uvicorn", "main:app", "--host", "0.0.0.0", "--port", "7860"]

requirements.txt

@@ -1,4 +1,4 @@
 fastapi
 uvicorn
-llama-cpp-python
+llama-cpp-python==0.3.2
 huggingface-hub