Dockerfile

# FROM python:3.11-slim

# # System dependencies - minimum tu
# RUN apt-get update && apt-get install -y \
#     gcc \
#     g++ \
#     && rm -rf /var/lib/apt/lists/*

# WORKDIR /app

# # Install llama-cpp-python KWANZA kwa pre-built wheel (haraka sana, no compile)
# RUN pip install --no-cache-dir \
#     llama-cpp-python==0.2.90 \
#     --extra-index-url https://abetlen.github.io/llama-cpp-python/whl/cpu \
#     --prefer-binary

# # Install dependencies zingine
# COPY requirements.txt .
# RUN pip install --no-cache-dir -r requirements.txt

# # Copy code
# COPY app/ ./app/

# # Tengeneza folders
# RUN mkdir -p ./data ./models

# EXPOSE 7860

# CMD ["uvicorn", "app.main:app", "--host", "0.0.0.0", "--port", "7860", "--workers", "1"]



FROM python:3.11-slim

# System dependencies
RUN apt-get update && apt-get install -y \
    gcc \
    g++ \
    cmake \
    libgomp1 \
    && rm -rf /var/lib/apt/lists/*

WORKDIR /app

# Install llama-cpp-python kutoka PyPI moja kwa moja (glibc compatible)
# --prefer-binary inachukua pre-built wheel ya glibc, si musl
RUN pip install --no-cache-dir \
    "llama-cpp-python==0.2.90" \
    --prefer-binary \
    --no-deps

# Install dependencies zingine
RUN pip install --no-cache-dir \
    diskcache \
    numpy

COPY requirements.txt .
RUN pip install --no-cache-dir -r requirements.txt

# Copy code
COPY app/ ./app/

# Tengeneza folders
RUN mkdir -p ./data ./models

EXPOSE 7860

CMD ["uvicorn", "app.main:app", "--host", "0.0.0.0", "--port", "7860", "--workers", "1"]