FROM python:3.13.7-slim

RUN python -m venv /opt/model-w \

&& /opt/model-w/bin/python -m pip install --no-cache-dir $MODEL_W_PIP_EXTRA 'modelw-docker[health]==2025.7.0' \

&& ln -s /opt/model-w/bin/modelw-docker /usr/local/bin/modelw-docker

RUN mkdir /code \

&& useradd -d /code -s /bin/bash user \

&& chown -R user:user /code

COPY --chown=user model-w.toml pyproject.toml poetry.lock README.md ./

ENV DEBIAN_FRONTEND=noninteractive

RUN apt-get update \

&& apt-get install -y curl ca-certificates gnupg apt-transport-https lsb-release \

&& curl https://www.postgresql.org/media/keys/ACCC4CF8.asc | gpg --dearmor --output /etc/apt/trusted.gpg.d/apt.postgresql.org.gpg \

&& curl https://deb.nodesource.com/gpgkey/nodesource.gpg.key | gpg --dearmor --output /etc/apt/trusted.gpg.d/nodesource.gpg

COPY ./sources.list /etc/apt/sources.list

RUN apt-get update \

&& apt-get install -y \

# Common project dependencies

libpcre3 \

libpcre3-dev \

mime-support \

postgresql-client \

libpq-dev \

gdal-bin \

gdal-data \

# Things required to build Python

build-essential \

gcc \

g++ \

make \

libssl-dev \

zlib1g-dev \

libbz2-dev \

libreadline-dev \

libsqlite3-dev \

llvm \

libncursesw5-dev \

xz-utils \

tk-dev \

libxml2-dev \

libxmlsec1-dev \

libffi-dev \

liblzma-dev \

# Things required to build Pillow

libtiff5-dev \

libjpeg-dev \

libopenjp2-7-dev \

zlib1g-dev \

libfreetype6-dev \

liblcms2-dev \

libwebp-dev \

tcl8.6-dev \

tk8.6-dev \

libharfbuzz-dev \

libfribidi-dev \

libxcb1-dev \

python3-dev \

# NodeJS

nodejs \

# Things required to use Weasyprint

libpango-1.0-0 \

libpangoft2-1.0-0 \

# Stuff for unstructured pdf parsing

poppler-utils \

# Stuff to allow configuration afterwards

sudo \

&& apt-get upgrade -y \

&& rm -fr \

/var/log/alternatives.log \

/var/log/apt/history.log \

/var/log/apt/term.log \

/var/log/bootstrap.log \

/var/log/dpkg.log \

/var/cache/man \

/var/cache/ldconfig/aux-cache \

/etc/machine-id \

/var/cache/apt

RUN echo "user ALL=(ALL:ALL) NOPASSWD: ALL" > /etc/sudoers.d/model-w \

&& chmod 0440 /etc/sudoers.d/model-w

RUN curl -sSL https://install.python-poetry.org | env POETRY_HOME=/opt/poetry python3 - \

&& ln -s /opt/poetry/bin/poetry /usr/local/bin/poetry

RUN modelw-docker install \

&& modelw-docker run poetry run playwright install-deps \

&& modelw-docker run poetry run playwright install firefox \

&& poetry run pip uninstall -y effdet

ARG INSTALL_CHAT_RAG

ENV INSTALL_CHAT_RAG=$INSTALL_CHAT_RAG

RUN if [ "$INSTALL_CHAT_RAG" = "true" ]; then echo "Installing nvidia-cudnn-cu12..."; pip install nvidia-cudnn-cu12==8.9.2.26; fi

RUN if [ "$INSTALL_CHAT_RAG" = "true" ]; then echo "Installing chat-rag..."; poetry run pip install chat-rag[full]==0.2.5; fi

RUN if [ "$INSTALL_CHAT_RAG" = "true" ]; then echo "Installing ninja-build..."; apt-get install ninja-build; fi

ENV RAY_task_events_max_num_task_in_gcs=100

CMD ./starter_script.sh