gpt-4o-mini-tts to gpt-realtime-2.1-mini
Some values not published Checked 2026-10-24.
Price per 1M tokens
| Field | Old | New | Change |
|---|---|---|---|
| Input | $0.6 | $0.6 | 0% (same) |
| Output | Not published | $2.4 | Not published |
| Cached input | Not published | $0.06 | Not published |
| Example monthly cost | Not published | $10.8 | Example only |
Context and max output
| Limit | Old | New | Change |
|---|---|---|---|
| Context window | Not published | 128,000 | Not published |
| Max output | Not published | 32,000 | Not published |
Capabilities that differ
No differing capability values are published for both models.
Endpoint changes
- The October 1, 2026 text-to-speech entry says these models are removed on January 6, 2027. Migrate to gpt-realtime-2.1-mini. See the Realtime API guide to plan your migration. Source
- The text-to-speech guide says this is a change of API workflow: gpt-realtime-2.1-mini is not a supported model for /v1/audio/speech. Existing integrations can use the Speech API (client.audio.speech.create, POST /v1/audio/speech) until shutdown. The Realtime API guide is https://developers.openai.com/api/docs/guides/realtime. Source
Checklist
- Read the shutdown date for gpt-4o-mini-tts on the linked official page.
- Point your requests at gpt-realtime-2.1-mini.
- Re-check price, context, and parameters on the provider pages linked here.
- Ship the change before the published shutdown date.