Skip to main content

Working with GPU's

Conveyor has support for GPU instances as documented in our Instances page.

We only support Nvidia GPU's.

We support the following methods of launching GPU instances:

Nvidia drivers​

Before you can use the Nvidia GPU's, you have to install the Nvidia drivers.

The easiest way to achieve this is to start from an Nvidia docker image:

FROM nvidia/cuda:12.9.0-cudnn-devel-ubuntu24.04

ARG PYTHON_VERSION="3.12"
ENV PYTHONUNBUFFERED=1

RUN apt-get update && \
    apt-get install -y python${PYTHON_VERSION} python3-venv python3-pip python-is-python3 && \
    rm -rf /var/lib/apt/lists/*

WORKDIR /app

RUN python -m venv /opt/venv
ENV PATH="/opt/venv/bin:$PATH"
ENTRYPOINT ["/opt/venv/bin/python"]

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

The previous example starts from a Nvidia image and installs your requirements and application code on top of it.

This will allow you to run the mnist example of pytorch