// AI Platform Drift
OpenAI /v1/audio/translations endpoint retires February 26, 2027; v1/realtime/translations is the replacement
OpenAI deprecations · published 2026-10-15 11:33 UTC
CORRECTED 2026-10-18 22:00 UTC — Correction, 18 October 2026: This item said both /v1/audio/translations and v1/realtime/translations retire on February 26, 2027. Only /v1/audio/translations has that shutdown date. OpenAI's deprecations table lists v1/realtime/translations as its recommended replacement. The February 26, 2027 date belongs to the 2026-08-26 transcription-models notice; it is not a follow-on to the January 20, 2027 removal of legacy audio, realtime and transcription models.
Stood 82h 26m · cause extraction error · every correction ↗
WHAT CHANGEDOpenAI added /v1/audio/translations to its 2026-08-26 transcription-models deprecation table, with a shutdown date of February 26, 2027 and v1/realtime/translations as the recommended replacement.
WHY IT MATTERSCode calling /v1/audio/translations will stop working on February 26, 2027.
WHAT TO DOSearch your codebase for /v1/audio/translations and plan the move to v1/realtime/translations before February 26, 2027.
Original text (as first published, superseded)
HEADLINEOpenAI audio translation endpoints retire February 26, 2027
WHAT CHANGEDOpenAI added February 26, 2027 as the retirement date for /v1/audio/translations and v1/realtime/translations endpoints. This is one month after the previously announced January 20, 2027 retirement of legacy audio, realtime, and transcription models.
WHY IT MATTERSAny code calling these translation endpoints will break after February 26, 2027. The staggered dates mean model retirement happens first (January 20) and endpoint removal follows (February 26), requiring two separate migrations if you use both deprecated models and translation endpoints.
WHAT TO DOGrep your codebase for /v1/audio/translations and v1/realtime/translations. If found, identify the replacement endpoints in OpenAI's current documentation and schedule migration work to complete before February 26, 2027.
Model-reported confidence 95% (not measured accuracy)
Primary source ↗
written by anthropic/claude-sonnet-4.5
Show the reasoning — raw diff, triage verdict, confidence
RAW DIFF
gpt-4o-transcribe-diarize gpt-live-transcribe or gpt-transcribe +Feb 26, 2027 +/v1/audio/translations +v1/realtime/translations + 2026-07-20: Legacy audio, realtime, and transcription models On July 20, 2026, we notified developers using legacy audio, realtime, and transcription model families and snapshots of their deprecation and removal from the API on January 20, 2027.
HOW IT WAS JUDGED
Triage signals below are model-reported assertions, preserved verbatim; they are not independent arithmetic verification.
| triage verdict | MATERIAL at 0.95, needed 0.45 |
|---|---|
| triage reasoning | New deprecation date and affected endpoint paths added to sunset schedule. |
| signals | future deprecation date added (Feb 26, 2027) · API endpoint sunset dates specified (/v1/audio/translations, v1/realtime/translations) · model removal timeline established |
| suppression | — |
| triage model | anthropic/claude-haiku-4.5 ($0.00000) |
| writer model | anthropic/claude-sonnet-4.5 ($0.00000) |