Clarification needed: GPT Standard deployments in Switzerland North

Michael Bock 5 Reputation points
2026-06-30T08:31:20.06+00:00

We are currently working on a production workload on Azure OpenAI in the Switzerland North region using any GPT Version that can be deployed with the Standard deployment type.

About one to two weeks ago, GPT-4.1 was available in the Switzerland North region with the Standard deployment type. However, this is no longer the case. Currently, no GPT large language models (LLMs) are available for Standard deployments in Switzerland North—only embedding and speech recognition models are listed.

Could you clarify when GPT LLMs will be available again in the Switzerland North region with the Standard deployment type?

Azure OpenAI in Foundry Models
0 comments No comments

2 answers

Sort by: Most helpful
  1. Michael Bock 5 Reputation points
    2026-06-30T15:23:05.8666667+00:00

    Those are the only models available when filtered Region Switzerland North, Deployment SKU Standard, Lifecycle Generally available. For DataZone Standard there are available gpt 5 series that would be fine, but for Deyployment SKU Standard there are only the listed model available .User's image

    Was this answer helpful?

    0 comments No comments

  2. Wang Wang 0 Reputation points
    2026-06-30T10:25:49.46+00:00

    Same issue here. According to the official docs, GPT-4o and GPT-4.1 shouldn't be deprecated until October this year — so either Microsoft moved the date up without telling anyone, or this is a bug.

    If the timeline was changed, the documentation needs to be updated immediately so people stop relying on wrong info. If it wasn't changed, then this is clearly a bug and it needs to be fixed ASAP.

    This is already seriously affecting our work. We shouldn't have to find out about critical changes through broken deployments instead of proper communication. Please fix this or update the docs — this isn't acceptable as it stands.

    Was this answer helpful?

    0 comments No comments

Your answer

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