what is the best choice for creating an Azure ML workspace, when there is the absolute need to keep the model and the code (mostly the model) proprietary?

Hi @Cherisse Robles,

Thank you for reaching out to Microsoft Q&A forum!

The best option for your Azure Machine Learning workspace depends on your specific requirements and constraints. Here's a brief overview of the options you've listed:

1. Public, restricted: This may be a good option if you need to collaborate with external partners or contractors, but still need to restrict access to your workspace.
2. Public, unrestricted: If you need to collaborate with a large team of contributors, but don't need to restrict access to your workspace.
3. Private, restricted: If you need to keep your workspace and data secure, but still need to collaborate with a small team of data scientists.
4. Private, unrestricted: If you need to keep your workspace and data secure, but still need to collaborate with a large team of contributors.

In general, if you need to keep your workspace and data secure, a private workspace with restricted access and a custom role for data scientists may be the best option. This option provides a secure environment with restricted access to authorized users only, and Private Link enabled ensures that access to the workspace is only through a private endpoint. The custom role for data scientists allows for more granular control over access to the workspace and its resources.

See: Private, Restricted workspace

I hope this information helps. Thank you.

If this answers your query, do click Accept Answer and Yes for was this answer helpful. And, if you have any further query do let us know.