Breaking model
OpenAI will remove the text-to-speech models tts-1, tts-1-hd and gpt-4o-mini-tts (with both its snapshots) from the API on January 6, 2027; the replacement is gpt-realtime-2.1-mini.
What changes
Announced October 1, 2026. OpenAI's deprecations page lists a January 6, 2027 shutdown for tts-1, tts-1-hd and gpt-4o-mini-tts with its snapshots gpt-4o-mini-tts-2025-03-20 and gpt-4o-mini-tts-2025-12-15, with gpt-realtime-2.1-mini as the replacement for each, and points to the Realtime API guide for speech migrations.
Breaking: Calls fail after this date.
What to do
Plan a move to gpt-realtime-2.1-mini through the Realtime API, following OpenAI's Realtime API guide. Its model page lists the Speech endpoint (v1/audio/speech) as not supported, so this is a code change, not a model-id swap; re-check voice output and cost.
Strings to search for
tts-1tts-1-hdgpt-4o-mini-ttsgpt-4o-mini-tts-2025-03-20gpt-4o-mini-tts-2025-12-15