# /v1/audio/translations shuts down on 2027-02-26

- **Provider:** OpenAI
- **Endpoint:** `POST /v1/audio/translations`
- **Effective:** 2027-02-26
- **Severity:** high
- **Change:** version-sunset
- **Status:** verified

## What changed

OpenAI deprecated the audio translations endpoint on 2026-08-26 and shuts it down on 2027-02-26, together with whisper-1 and the gpt-4o transcribe models. After that date every call to POST /v1/audio/translations fails.

## What to do

Replace calls to POST /v1/audio/translations with the Realtime translations interface (v1/realtime/translations) that OpenAI names as the replacement, and move transcription off whisper-1 to gpt-live-transcribe or gpt-transcribe before 2027-02-26.

## Sources

- Provider notice (2026-08-26): https://developers.openai.com/api/docs/deprecations
- https://developers.openai.com/api/reference/resources/audio

## Machine-readable

- Advisory: https://github.com/SriVarshini09/shapi-advisories/blob/main/advisories/openai-audio-translations-endpoint-shutdown.json
- OSV: https://shapi.dev/osv/openai-audio-translations-endpoint-shutdown.json