Unsupported Base Models in Azure AI Foundry for West Europe Region

Aleksa Janjić 0 Reputation points
2025-04-24T11:58:50.88+00:00

A problem is being encountered with the Azure AI Foundry service regarding model availability in the West Europe region.

Two separate Azure subscriptions have been created within the organization, along with different AI hubs and AI projects in the West Europe region. The issue arises from one project having several different models deployed as serverless APIs, while the other project does not support any models for deployment in the same region. For example, the first project allows for the deployment of DeepSeek or Llama models without limitations, whereas the second project faces restrictions that prevent the deployment of these models due to unavailability in the West Europe region.

Azure OpenAI Service
Azure OpenAI Service
An Azure service that provides access to OpenAI’s GPT-3 models with enterprise capabilities.
3,976 questions
{count} votes

1 answer

Sort by: Most helpful
  1. SriLakshmi C 4,880 Reputation points Microsoft External Staff Moderator
    2025-04-24T17:43:32.6066667+00:00

    Hello Aleksa Janjić,

    I understand that you're encountering a model availability issue within Azure AI Foundry in the West Europe region. Although both of your projects are set up in the same region, the difference in available models such as DeepSeek and Llama can be attributed to a few factors.

    Azure AI Foundry’s support for models can vary not only by region but also by subscription tier, project configuration, and rollout stage. While one project may have access to deploy certain models due to early access, specific settings, or subscription eligibility, the other might not support them yet due to these constraints. It’s also worth noting that some models may have deployment restrictions tied to resource availability or compliance requirements in specific regions.

    To resolve this, you should verify if the models in question are officially supported in West Europe by checking Azure’s documentation. Additionally, confirm whether both subscriptions and AI hubs are using the same tier and configurations. If everything appears aligned but the issue persists, it could be due to a phased rollout or internal permissions.

    Kindly refer this Region availability for models in serverless API endpoints.

    I Hope this helps. Do let me know if you have any further queries.


    If this answers your query, please do click Accept Answer and Yes for was this answer helpful.

    Thank you!

    0 comments No comments

Your answer

Answers can be marked as Accepted Answers by the question author, which helps users to know the answer solved the author's problem.