Model retirement dates / Azure

gpt-4o-transcribe on Azure

On Azure, gpt-4o-transcribe retired on 15 October 2026 (version 2025-03-20). On OpenAI's own API the date is different: retires 26 February 2027.

PlatformMicrosoft Azure (Azure OpenAI / Foundry Models)
Modelgpt-4o-transcribe by OpenAI
StatusRetired (as of 16 Nov 2026)
Retirement15 October 2026
On OpenAI's APIgpt-4o-transcribe · retires 26 Feb 2027
Model typeAudio
SourceMicrosoft: model retirement schedule

Dates by version

Azure retires each model version on its own date. Check which version your deployment uses (deployment details in the Azure portal, or version in your Bicep or Terraform).

VersionLifecycleRetiresReplacement
2025-03-20GA15 Oct 2026—

How Driftline tells Azure from OpenAI's API

Azure uses OpenAI's model names. Driftline uses Azure's dates in files that show they call Azure: AzureOpenAI, azure_endpoint, openai.azure.com, AZURE_OPENAI_…, azure/ model names, and Bicep or Terraform deployments, where it also reads the model version (never api_version). Elsewhere it uses OpenAI's dates and notes when Azure's differ.

Check your code for it

Run this in your repository. No install and no API key; it lists every OpenAI, Anthropic, Google, Mistral, xAI, Cohere, DeepSeek, Groq, Bedrock, Vertex AI and Azure model ID it finds, with its retirement date and replacement:

npx -y @aidriftline/driftline .

Or look for this ID only: grep -rn "gpt-4o-transcribe" . (misses aliases, Bedrock and Vertex names, and IDs split across lines).

To be warned automatically before the next retirement, add the free Driftline GitHub Action. It annotates the exact line in pull requests and can fail the build on retired models.

Where a new name is the whole change, Driftline can make it for you: npx -y @aidriftline/driftline fix --dry-run shows the edits, and the Action's fix-pr: true opens one pull request with them, plus a checklist for the lines that need more. See an example pull request.

Read from Microsoft's official schedule on 16 November 2026, checked every 6 hours. Something wrong? hello@aidriftline.com