# whisper-1

Calls to the model whisper-1 on OpenAI stop working on February 26, 2027.

On February 26, 2027, OpenAI will shut down the whisper-1 model. Calls targeting model whisper-1 will fail after that date.

## The error OpenAI returns once whisper-1 is gone

HTTP 404

```json
{"error":{"message":"The model `whisper-1` has been deprecated, learn more here: https://platform.openai.com/docs/deprecations","type":"invalid_request_error","param":null,"code":"model_not_found"}}
```

## What to use instead

Move to gpt-live-transcribe or gpt-transcribe. Switching to gpt-live-transcribe requires more than changing an identifier, as it only runs in Realtime transcription sessions (/v1/realtime/transcription_sessions) where audio is streamed in and transcript deltas return, rather than uploading files to /v1/audio/transcriptions.

gpt-live-transcribe only runs in Realtime transcription sessions (/v1/realtime/transcription_sessions): audio is streamed in and transcript deltas come back, with no file upload to /v1/audio/transcriptions.

- API: OpenAI (api.openai.com)
- Model: whisper-1
- Released: March 1, 2023 (models.dev)
- Kind: Model retirement
- Takes effect: February 26, 2027
- Calls after that day: fail
- Calls it hits: `model whisper-1`
- Source: [Deprecations page](https://developers.openai.com/api/docs/deprecations)
- Verified: November 6, 2026, against OpenAI's own pages
- Page: https://www.okapi.bot/models/openai/whisper-1/

## OpenAI's words

> Shutdown date: Feb 26, 2027; Model / system: whisper-1; Recommended replacement: gpt-live-transcribe or gpt-transcribe

Okapi finds the calls this change hits in your repo and opens the pull request with the fix: https://github.com/apps/okapi-agent