Dockerfile

FROM nvidia/cuda:12.3.2-cudnn9-devel-ubuntu22.04

# Set environment variables
ENV PYTHONUNBUFFERED=1 \
    DEBIAN_FRONTEND=noninteractive \
    CUDA_HOME=/usr/local/cuda \
    PATH=/usr/local/cuda/bin:$PATH \
    LD_LIBRARY_PATH=/usr/local/cuda/lib64:$LD_LIBRARY_PATH \
    NVIDIA_VISIBLE_DEVICES=all \
    NVIDIA_DRIVER_CAPABILITIES=compute,utility \
    HF_HOME=/app/models \
    TRANSFORMERS_CACHE=/app/models \
    TRITON_CACHE_DIR=/tmp/triton_cache \
    XDG_CACHE_HOME=/tmp \
    NUMBA_CACHE_DIR=/tmp/numba_cache

# Install system dependencies
RUN apt-get update && apt-get install -y --no-install-recommends \
    software-properties-common \
    && add-apt-repository ppa:deadsnakes/ppa \
    && apt-get update \
    && apt-get install -y --no-install-recommends \
    python3.10 \
    python3.10-pip \
    python3.10-dev \
    python3.10-distutils \
    build-essential \
    git \
    ffmpeg \
    libsndfile1 \
    curl \
    && rm -rf /var/lib/apt/lists/*

# Create symlinks for python3.10
RUN ln -sf /usr/bin/python3.10 /usr/bin/python3 \
    && ln -sf /usr/bin/python3.10 /usr/bin/python

# Upgrade pip and install build tools
RUN python3 -m pip install --upgrade pip setuptools wheel uv

WORKDIR /app

# Create cache directories and set permissions
RUN mkdir -p /tmp/numba_cache /tmp/triton_cache /app/models && \
    chown -R nobody:nogroup /tmp/numba_cache /tmp/triton_cache /app/models && \
    chmod -R 755 /tmp/numba_cache /tmp/triton_cache /app/models

COPY requirements.txt .

# Install other requirements
RUN python3 -m uv pip install -r requirements.txt --prerelease=allow
RUN python3 -m nltk.downloader punkt punkt_tab

COPY . .

EXPOSE 8000

CMD ["python3", "newstartup.py"]