For AI agents: the complete documentation index is available at https://docs.us.ovhcloud.com/en/llms.txt, the full documentation bundle is available at https://docs.us.ovhcloud.com/en/llms-full.txt, and this page is available as Markdown at https://docs.us.ovhcloud.com/en/public-cloud/ai-machine-learning/ai-notebooks-capabilities.md.
  • English

AI Notebooks - Features, capabilities, and limitations

View as Markdown

Learn about the technical features, capabilities, and limitations of the Public Cloud AI Notebooks offer.

Info

AI Notebooks is covered by OVHcloud Public Cloud Special Conditions.

 

Features

Available features

AI Notebooks are Managed Jupyter or VS Code notebooks linked to compute resources (CPUs/GPUs) and storage. You can compare them to Google Colab or Amazon Sagemaker notebooks.

FeatureDetails
Live code editor and AI environments
Jupyter and VS CodeYou can use Jupyter or VS Code as your preferred live-code editor. If you opt for VS Code, you can also set up a remote connection (for example, from your laptop).
Preinstalled Machine Learning environmentsAI Notebooks comes with a generic Python environment (Conda) or pre-installed ones, such as Pytorch, Tensorflow, HuggingFace, and more.
Easy customizationAI Notebooks allows installation of almost any Conda or Pip packages. You can easily customize your environment to suit your needs.
Management
Multiple ways to manage your notebooksYou can manage your AI Notebooks through the OVHcloud Control Panel, CLI, API or Python SDK. Depending on your needs, you can easily automate their creation and deletion as well.
Easy start and StopYou can start and stop a notebook in one click. Once stopped, your notebook environment is kept, and you can restart it later without losing your data and experiments.
Compute resources
Guaranteed compute resourcesSelect the number of CPUs or GPUs required during the creation of the AI Notebooks. Once launched, you will keep these resources as long as your notebook is running.
Background executionYour tasks can be executed in the background, meaning that closing your web browser will not affect your work.
No maximum runtimeYour tasks can last as long as your notebook is running.
Monitoring toolsEach AI Notebooks service comes with a native Grafana dashboard, allowing you to keep track of and monitor your CPU, GPU, RAM, and storage resources.
Storage
Fast and flexible storageEach AI Notebooks service comes with local storage, but also the ability to attach remote storage from Object Storage. From a few GiB to multiple TiB, we push your data near our compute power on fast SSD storage for better performance.
Git repositories importationDuring the creation of your AI Notebooks, you can specify one or multiple Git repositories to download inside your notebook environment.
Security
Open or restricted authenticationDuring the creation of your AI Notebooks, select open or restricted access to your notebook. If restricted, people can be granted access via token or credentials to securely access your environment.
Billing
Easy billingYou only pay for what you consume, billed per minute with per-second granularity.

 

Command line interface (CLI)

AI Notebooks is compliant with the OVHcloud AI CLI. Discover how to install the OVHcloud AI CLI.

 

Monitoring tools

To see information about your notebook, you can do so with the ovhai CLI using this command:

ovhai notebook get <notebook-id>

You can access your metrics through the Monitoring Url.

You are also able to check it from the OVHcloud Control Panel in your notebook's general information by clicking the Go to Graph Dashboard button.

 

Planned features

We continuously improve our offers. You can follow, vote, and submit ideas to add to our roadmap.

 

Capabilities and limitations

Supported regions

AI Notebooks can be used from any country in the world, as long as you have an OVHcloud account. Our AI & Machine Learning services are based in the US-EAST-VA (Vint Hill, Virginia) region.

 

Attached resources

Compute resources

You can choose the number of GPUs or CPUs for a notebook, but cannot use both. By default, a notebook uses one GPU. The memory resource is not customizable.

If you choose GPU:

  • CPU, memory, and local storage resources are not customizable but scaled linearly with each additional GPU.

If you choose CPU:

  • Memory and local storage resources are not customizable but scaled linearly with each additional CPU.

The maximum amount of CPU/GPU, memory per CPU/GPU, and local storage is available on the OVHcloud website, Control Panel, and the ovhai CLI.

ovhai capabilities flavor list

For your information, the current limits are:

  • CPU: 12 per notebook.
  • GPU: 4 per notebook.

 

Available hardware for AI Notebooks

Currently, we provide:

  • ai1-1-cpu
  • L4-1-gpu
  • L40s-1-gpu

 

Available storage

Local storage

Each AI Notebook comes with a local storage space, which is ephemeral. When you delete your notebook, this storage space is also deleted. This storage space depends on the selected instances during the notebook creation. Please refer to the compute resources section for more information.

Info

Local storage is limited and not the recommended way to handle data, see the OVHcloud documentation on data for more information.

 

Attached storage

When attaching data volumes to your AI Notebooks, you can use storage from Public Cloud Object Storage. This allows you to work with large datasets, while ensuring persistence even if you delete your notebooks.

Info

For best performance, we recommend keeping the Object Storage bucket in the same region as your AI Notebook. Mounting a bucket from a different region also works, but adds cross-region network latency. If you use an external S3* datastore registered via the ovhai CLI with --store-credentials-locally, keep in mind that the datastore is only registered for the CLI region context (ovhai config set <region>) it was created under, so you need to register it again under each region you use it from.

Your Public Cloud Project can store unlimited data in Object Storage buckets. However, when you mount an Object Storage bucket as a volume, there is a usage limit of 10 TB per Public Cloud Project. This limit applies to the total storage consumed by all volumes attached simultaneously across your AI Notebooks, AI Training jobs, and AI Deploy apps.

When the same volume is used across multiple resources, enabling caching allows shared access to the volume data, preventing multiple copies from consuming additional storage quota. Without caching, each instance will maintain a separate copy of the volume data, increasing the total storage usage linearly.

 

Maximum execution time

Warning

NOTE: There is no duration limitation on AI Notebooks execution.

However, by default, your AI Notebook will automatically shut down after 7 consecutive days of being in a RUNNING state. All your settings and data are preserved and you can start it again anytime. You can also enable Automatic Restart to have it restart every 7 days.

 

Live-code editors

You can choose between two live-code editors to launch and edit your notebook:

  • Jupyterlab
  • VS Code

You cannot install your own code editor on AI Notebooks.

With VS Code, you get the capability to use remote connections (from a local computer).

 

Pre-installed AI environments

OVHcloud AI Notebooks come with pre-installed AI environments.

List of available AI Environments:

  • AutoGluon + MXNet
  • FastAI
  • HuggingFace Transformers
  • JAX
  • Miniconda (Python generic)
  • Miniconda with Colab Compatibility
  • MLR3
  • MXNet
  • One image to rule them all
  • PyTorch
  • Scikit-Learn
  • TensorFlow

 

Environment customization

Each environment can be customized directly with PIP or CONDA (we support almost any package and library).

Limitations:

  • You are not the administrator (root). You cannot install Linux packages (such as apt-get).
  • AI Notebooks does not allow the use of custom Docker images. In case you need a very specific package or framework, you can bring your custom Docker images with OVHcloud AI Training.

AI Training allows you to benefit from the same technology and pricing, but you can create notebooks directly with your own Docker images. If you want to build and use a custom Docker image, you can do it with AI Training by following this tutorial.

 

Network

  • Public networking can be used for all the AI Tools.
  • Private networking (OVHcloud vRack) is not supported.

 

Available ports to public network

Each notebook has a public URL that accesses port 8080 of the notebook by default and cannot be changed.

Notebook URL for accessing the default port (starting with the notebook's ID):

However, you can access other ports than the default 8080 by appending its number to your notebook URL. For example, if you want to access port 8501 from your notebook, you will have to add -8501 to your notebook URL. Your notebook URL for accessing the 8501 port will then be https://00000000-0000-0000-0000-000000000000-8501.notebook.us-east-va.ai.cloud.ovh.us.

Only the HTTP layer is accessible.

 

Quotas per Public Cloud project

GPU and CPU availability is subject to the quota set on your Public Cloud Project. You can reach out to our support team to increase your quota.

 

Go further

Browse all the AI & Machine Learning guides.

Browse the rest of the OVHcloud US documentation for guides on our other products and services.

Learn more about AI & Machine Learning on the OVHcloud US website.

If you need training or technical assistance to implement our solutions, contact your sales representative or click on this link to get a quote and ask our Professional Services experts for a custom analysis of your project.

*S3 is a trademark filed by Amazon Technologies, Inc. OVHcloud's service is not sponsored by, endorsed by, or otherwise affiliated with Amazon Technologies, Inc.