GPT 4 models are deprecated and I'm unable to deploy equivalent GPT 5 models in my region

Hello Mikko,

Greetings! Thanks for raising this question in the Q&A forum

I checked the current region availability tables directly, and your observation is accurate. As of the latest published documentation, the Data Zone Batch deployment type for the Europe geography only lists gpt-4.1, gpt-4.1-mini, gpt-4.1-nano, gpt-4o, gpt-4o-mini, o3, and o4-mini. GPT-5 and GPT-5.1 are present in Data Zone Batch for the Americas geography, and they are present in the EU Data Zone for the Standard (non-batch) deployment type, but they have not yet been rolled out to Data Zone Batch specifically for Europe. So this isn't a configuration mistake on your side, it's a genuine gap in regional/deployment-type rollout that happens to line up badly with the GPT-4 retirement timeline.

1. Confirm this is the actual gap you're hitting. You can verify the current state yourself at any time, since Microsoft updates this table as rollout progresses:

      https://learn.microsoft.com/en-us/azure/foundry/foundry-models/concepts/models-sold-directly-by-azure-region-availability?pivots=batch&tabs=az-europe
   

   Scroll to the "Data Zone batch" section and check the Europe tab specifically, since GPT-5 could be added there without much notice as Microsoft continues expanding capacity.
2. Consider switching to Data Zone Standard (non-batch) for GPT-5 while staying in the EU zone. This is the most direct workaround that preserves your "data stays in Europe" requirement, since GPT-5 and GPT-5.1 are both already available under Data Zone Standard for Europe. This changes your architecture from asynchronous batch submission to synchronous pay-per-token calls, so if your workload can tolerate that (or you can build your own lightweight queuing/throttling layer on top of the standard endpoint to emulate batch-style processing), this unblocks you immediately without waiting on Microsoft's rollout.
3. Alternatively, use a Regional (single-region) Standard deployment in an EU region if batch specifically isn't required. GPT-5 is available as a Standard/Regional deployment in uksouth today. This pins your processing to that one specific region rather than the wider EU data zone, which is a stricter form of the same guarantee, but it does not currently extend to other EU regions like francecentral or westeurope for GPT-5 at Standard/Regional tier, only uksouth. Confirm whether UK region residency satisfies your specific compliance requirement before choosing this path, since UK and EU are distinct data-protection jurisdictions.
4. If your workload genuinely requires Batch specifically within the EU data zone, open a support case citing the deprecation timeline conflict. Since GPT-4 retirement removes your only currently compliant EU Data Zone Batch option without Microsoft having yet shipped the GPT-5 equivalent in that same deployment type and geography, this is worth raising as a business-impact case rather than a general "how do I" question:

      Azure Portal > Help + Support > Support + Troubleshooting > Service and subscription limits (quotas) or Technical > Microsoft Foundry
   

   Reference the specific gap (GPT-5/5.1 present in Americas Data Zone Batch and EU Data Zone Standard, but absent from EU Data Zone Batch) and ask whether an accelerated rollout timeline or a temporary retirement extension is possible for your subscription while the gap remains open.
5. Track the model retirement schedule for updates. Microsoft periodically adjusts retirement dates when replacement parity issues like this are raised at scale, so it is worth monitoring:

      https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule
   

If this answer helps you kindly accept the answer which will help others who have similar questions.

Best Regards,

Jerald Felix.