OpenAI · model
gpt-4o-mini-realtime-preview deprecation: shutdown date & replacement
- Provider
- OpenAI
- Type
- model
- Announced
- Shutdown
- Replacement
gpt-realtime-mini- Source
- Official OpenAI notice
Page says notified in September 2025; section dated 2025-09-15.
How to migrate off gpt-4o-mini-realtime-preview
- Find every place that references it:
grep -rn --exclude-dir={node_modules,.git,dist} "gpt-4o-mini-realtime-preview" . - Replace the model ID:
- model: "gpt-4o-mini-realtime-preview" + model: "gpt-realtime-mini" - Re-run your evals or a sample of real prompts — output style, token usage and pricing can change between models.
- Check config outside code too: environment variables, feature flags, prompt-management tools, and fine-tuned model references.
- Deploy before May 7, 2026, and add the CI scanner so it cannot come back.
Also shutting down on May 7, 2026 (OpenAI)
gpt-4o-audio-previewgpt-4o-mini-audio-previewgpt-4o-realtime-previewgpt-4o-realtime-preview-2024-12-17gpt-4o-realtime-preview-2025-06-03
Questions
Is gpt-4o-mini-realtime-preview deprecated?
Yes. OpenAI has deprecated gpt-4o-mini-realtime-preview. It was shut down on May 7, 2026.
When does gpt-4o-mini-realtime-preview shut down?
May 7, 2026, according to OpenAI's official deprecation notice.
What should I use instead of gpt-4o-mini-realtime-preview?
OpenAI recommends gpt-realtime-mini. Switch the model ID and re-test your prompts, since outputs, pricing and limits can differ.
What happens to requests to gpt-4o-mini-realtime-preview after the shutdown?
Requests using the retired model ID fail with an error, so any code still calling it breaks. Update and deploy before the shutdown date.