1Dockerfile

FROM ubuntu:22.04

# Install dependencies
RUN apt-get update && apt-get install -y \
    git \
    build-essential \
    cmake \
    curl \
    python3-pip \
    && rm -rf /var/lib/apt/lists/*

# Install llama.cpp
RUN cd / && git clone https://github.com/ggerganov/llama.cpp  && cd llama.cpp && make

# Setup model directory
WORKDIR /models

# Python dependencies
COPY requirements.txt .
RUN pip install -r requirements.txt

# Copy app files
COPY app.py .

# Download model and start server
CMD ["bash", "-c", "cd /models && python3 app.py"]