GPT-Audio
GPT-Audio is a deprecated audio model accessed through Chat Completions. It handles text and audio without requiring the Realtime session interface.
Deprecated API; check access details
A REST audio model with its own history
The default dated snapshot is from August 2025. The model can receive and produce text or audio, with a 128,000-token window and a 16,384-token output ceiling. It is distinct from the later GPT-Audio-1.5 release.GPT-Audio official model documentation
Review an existing integration
Practical guidance. When updating a voice-note feature, compare the accepted audio formats, returned content and tool-call behavior. Preserve the request-response interaction if that is what the product needs. A Realtime replacement is a different integration decision.
Specifications & access
| Detail | Value and conditions |
|---|---|
| Status | Deprecated API; check access details |
| Access | API |
| Model identifiers | gpt-audiogpt-audio-2025-08-28 |
| Context window | 128,000 tokens Published context for this deprecated release GPT-Audio official model documentation |
| Maximum output | 16,384 tokens Published maximum output, bounded by the remaining context GPT-Audio official model documentation |
| Provider status | Deprecated in the current OpenAI model catalog; confirm remaining account access.GPT-Audio official model documentation |
Related models
- GPT-Audio-1.5
OpenAI ยท GPT Audio
GPT-Audio-1.5 works with audio through Chat Completions.
Launch date not confirmedAPI