Skip to content

Workbenches

aTheo edited this page Oct 10, 2025 · 17 revisions

Workbench images are supported for a minimum of one year. Major updates to pre-configured notebook images occur approximately every six months. Therefore, two supported notebook images are typically available at any given time. If necessary, you can still access older notebook images from the registry, even if they are no longer supported. You can then add the older notebook images as custom notebook images to cater to the project’s specific requirements.

Open Data Hub contains the following workbench images with different variations:

Workbench Version Python OS x86_64 aarch64 ppc64le s390x Container
CPU Minimal - 3.12 UBI9 quay.io/repository/opendatahub/odh-workbench-jupyter-minimal-cpu-py312-ubi9
CPU DataScience - 3.12 UBI9 quay.io/repository/opendatahub/odh-workbench-jupyter-datascience-cpu-py312-ubi9
CPU TrustyAI - 3.12 UBI9 - quay.io/repository/opendatahub/odh-workbench-jupyter-trustyai-cpu-py312-ubi9
CPU CodeServer - 3.12 UBI9 quay.io/repository/opendatahub/odh-workbench-codeserver-datascience-cpu-py312-ubi9
CPU RStudio - 3.12 C9S - - quay.io/repository/opendatahub/odh-workbench-rstudio-minimal-cpu-py312-c9s
CUDA Minimal 12.8 3.12 UBI9/RHEL9.6 quay.io/repository/opendatahub/odh-workbench-jupyter-minimal-cuda-py312-ubi9
CUDA PyTorch 12.8 3.12 UBI9/RHEL9.6 quay.io/repository/opendatahub/odh-workbench-jupyter-pytorch-cuda-py312-ubi9
CUDA PyTorch LLMCompressor 12.8 3.12 UBI9/RHEL9.6 quay.io/repository/opendatahub/odh-workbench-jupyter-pytorch-llmcompressor-cuda-py312-ubi9
CUDA TensorFlow 12.8 3.12 UBI9/RHEL9.6 quay.io/repository/opendatahub/odh-workbench-jupyter-tensorflow-cuda-py312-ubi9
CUDA RStudio 12.8 3.12 C9S quay.io/repository/opendatahub/odh-workbench-rstudio-minimal-cuda-py312-c9s
ROCM Minimal 6.3 3.12 UBI9/RHEL9.6 quay.io/repository/opendatahub/odh-workbench-jupyter-minimal-rocm-py312-ubi9
ROCM PyTorch 6.3 3.12 UBI9/RHEL9.6 quay.io/repository/opendatahub/odh-workbench-jupyter-pytorch-rocm-py312-ubi9
ROCM TensorFlow 6.3 3.12 UBI9/RHEL9.6 quay.io/repository/opendatahub/odh-workbench-jupyter-tensorflow-rocm-py312-ubi9
Runtime Version Python UBI x86_64 aarch64 ppc64le s390x Container
CPU Minimal - 3.12 UBI9 quay.io/repository/opendatahub/odh-pipeline-runtime-minimal-cpu-py312-ubi9
CPU DataScience - 3.12 UBI9 quay.io/repository/opendatahub/odh-pipeline-runtime-datascience-cpu-py312-ubi9
CUDA PyTorch 12.8 3.12 UBI9/RHEL9.6 quay.io/repository/opendatahub/odh-pipeline-runtime-pytorch-cuda-py312-ubi9
CUDA PyTorch LLMCompressor 12.8 3.12 UBI9/RHEL9.6 quay.io/repository/opendatahub/odh-pipeline-runtime-pytorch-llmcompressor-cuda-py312-ubi9
CUDA TensorFlow 12.8 3.12 UBI9/RHEL9.6 quay.io/repository/opendatahub/odh-pipeline-runtime-tensorflow-cuda-py312-ubi9
ROCM PyTorch 6.3 3.12 UBI9/RHEL9.6 quay.io/repository/opendatahub/odh-pipeline-runtime-pytorch-rocm-py312-ubi9
ROCM TensorFlow 6.3 3.12 UBI9/RHEL9.6 quay.io/repository/opendatahub/odh-pipeline-runtime-tensorflow-rocm-py312-ubi9
  • ✅ supported
  • ⚠️ under development
  • ⏰ planned in the future
  • ❌ not supported by vendor

These notebooks are incorporated to be used in conjunction with Open Data Hub, specifically utilizing the ODH Notebook Controller as the launching platform. The table above provides insights into the characteristics of each notebook, including their availability in both ODH and OpenShift AI environments, GPU support, and whether they are offered as runtimes ie without the JupyterLab UI.

All the notebooks are available on the Quay.io registry; please filter the results by using the tag "2023b" for the latest release and "2023a" for the n-1.

Jupyter Minimal

Jupyter Minimal provides a browser-based integrated development environment where you can write, edit, and debug code using the familiar interface and features of JupyterLab. If you do not require advanced machine learning features or additional resources for compute-intensive data science work, you can use the Minimal Python image to develop your models.

CUDA

CUDA provides a browser-based integrated development environment where you can write, edit, and debug code using the familiar interface and features of JupyterLab. If you are working with compute-intensive data science models that require GPU support, use the Compute Unified Device Architecture (CUDA) notebook image to gain access to the NVIDIA CUDA Toolkit. You can optimize your work using GPU-accelerated libraries and optimization tools using this toolkit.

Jupyter Data Science

Standard Data Science provides a browser-based integrated development environment where you can write, edit, and debug code using the familiar interface and features of JupyterLab. Use the Standard Data Science notebook image for models that do not require TensorFlow or PyTorch.
This image contains commonly used libraries to assist you in developing your machine-learning models. Furthermore, we have integrated several useful libraries and applications. Notably, we've included Mesa-libgl, an additional library designed for OpenCV tasks. We've also introduced Git-lfs, which provides an efficient solution for handling large files, such as audio samples, videos, datasets, and graphics. The integration of unixODBC offers a standardized API for accessing data sources, including SQL Servers and other data sources with ODBC drivers. Lastly, the addition of Libsndfile makes it easier to read and write files containing sampled audio data. Additionally, this notebook comes equipped with standard database clients for MySQL, PostgreSQL, MSSQL, and MongoDB.

NOTE: All notebooks derived from the Jupyter Data Science Notebook inherit these libraries and applications, with the exception of the minimal and CUDA variants.

Jupyter Tensorflow

TensorFlow is an open-source platform for machine learning. It provides a browser-based integrated development environment where you can write, edit, and debug code using the familiar interface and features of JupyterLab. With TensorFlow, you can build, train and deploy your machine learning models. TensorFlow contains advanced data visualization features, such as computational graph visualizations. It also allows you to easily monitor and track the progress of your models.

Jupyter PyTorch

PyTorch is an open-source machine learning library optimized for deep learning. If you are working with computer vision or natural language processing models, use the Pytorch notebook image. It provides a browser-based integrated development environment where you can write, edit, and debug code using the familiar interface and features of JupyterLab.

Jupyter PyTorch & LLM Compressor

PyTorch and LLM Compressor have been combined to help you efficiently optimize and deploy your models with vLLM, a high-performance library for LLM inference and serving. Use the PyTorch notebook image to develop and fine-tune models—ideal for computer vision and natural language processing tasks. It provides a browser-based JupyterLab environment where you can easily write, edit, and debug your code.

Jupyter TrustyAI

Use the TrustyAI notebook image to leverage your data science work with model explainability, tracing and accountability, and runtime monitoring. It provides a browser-based integrated development environment where you can write, edit, and debug code using the familiar interface and features of JupyterLab.

Code Server

Code Server (VS Code) provides a browser-based integrated development environment (IDE) where you can write, edit, and debug code using the familiar interface and features of VS Code. It is particularly useful for collaborating with team members, as everyone can access the same development environment from their own devices.

R Studio

It provides a powerful integrated development environment specifically designed for R programming. By integrating R Studio IDE into ODH, you equip data analysts with a dedicated environment for exploring and manipulating data, building models, and generating insightful visualizations. Moreover, If you are working with compute-intensive data science models that require GPU support, use the CUDA R Studio notebook image to gain access to the NVIDIA CUDA Toolkit.

Previous Page | Next Page

Clone this wiki locally