Thank you for the detailed analysis and guidance.
I was able to resolve the issue.
Although I had deployed the model following the code-based approach recommended in the Hugging Face documentation, the deployment was not being reflected correctly in my environment due to an internal corporate policy restriction. Once that policy-related issue was identified and addressed, the deployment completed successfully and the model became available as expected.
Your breakdown of the potential failure points was very helpful and aligns with the areas I initially investigated. Fortunately, the endpoint identity, workspace connection, and HF token configuration were not the source of the problem in this case.
Thanks again for taking the time to provide such a thorough troubleshooting path—it will certainly be useful for anyone encountering similar symptoms in Azure AI Foundry/Azure ML deployments.