GPT Audio
OpenAI · released Jan 19, 2026
Speech to speechTakes speech and answers in speech
- Type
- Proprietary
- Input
- $2.50
- Output
- $10.00
- Cached
- Not published
List price · per 1M tokens · OpenAI at 128K context · source ↗
Our take
Written Aug 3, 2026GPT Audio is OpenAI's proprietary speech-to-speech model launched in January 2026, handling audio and text in both directions with a 128,000-token request limit. It is an ecosystem bet rather than a measured pick, since no accuracy, speed or language coverage data is available.
Choose this for real-time speech-to-speech applications where OpenAI ecosystem integration matters, or for voice interactions that need an unusually long context buffer. Skip it if you need verified accuracy, speed benchmarks, or language coverage before committing to a hosted-only model.
The case for it
- 128,000-token request limit far exceeds typical speech-to-text buffers, enabling long-form audio sessions or document-plus-voice workflows.
- Hosted by OpenAI directly and through OpenRouter, giving two provider options.
The case against it
- No verified accuracy or speed data, so performance against any task is unassessed.
- Hosted only with no measured quality to justify its position in the price range.
How good is it?
We hold no score for this model.
The Open ASR Leaderboard is where a model that turns speech into text gets measured, and GPT Audio is not on it — so we hold no accuracy figure for it, rather than a poor one.
Or rent it from someone else
Why this differs from the header. The strip above quotes OpenAI's own list price. This is the cheapest live offer at the widest standard context we hold, whoever is serving it — a reseller undercutting a lab is ordinary commerce, not an error.
- per 1M tokens
- $2.50 in / $10.00 out
- Context served
- 128K
- Throughput
- Not measured
| Provider | In / out per 1M tokens | Context | Throughput | Trains on prompts | Logs prompts | Zero retention |
|---|---|---|---|---|---|---|
| OpenRouter | $2.50 / $10.00 | 128K | not measured | Unknown | Unknown | Unknown |
| OpenAI | $2.50 / $10.00 | 128K16K out | 71 tok/s | No | Yesunknown period | Unknown |
Across the 2 listings we hold: 1 say they do not train on prompts, 0 say they do and 1 do not say. 0 appear in the zero-retention registry we check; the rest are unknown to us rather than confirmed either way.
What each host's API supports
From the parameter list each endpoint publishes. Streaming is omitted: nothing we hold reports it, for any model.
- ✓
- Supported
- ✗
- Not supported
- Not published
- host gave no parameter list
| Provider | Tool calling | JSON output | Strict schema |
|---|---|---|---|
| OpenRouter | ✓ | ✓ | ✓ |
| OpenAI | ✓ | ✓ | ✓ |
Tool calling: 2 of 2 listings say yes. JSON output: 2 of 2 listings say yes. Strict schema: 2 of 2 listings say yes.
When we formed this view
Dates behind this page
Prices last checked 9d ago
What we do not know about this model yet
- No board we watch has turned up a score, so we hold no quality figures at all.
- Nothing we hold says whether an endpoint streams, so we do not show it either way.
- 1 of 2 listings do not say whether they train on prompts.
- We hold no cached-input rate for any of its listings.
Licence and identifiers
What the licence allowsWe hold no licence record for this model. Inside are the identifiers you need to pull it — its Hugging Face repo where we have one, our slug and a machine-readable card.
Licence
Commercial API terms. We hold no licence record for this model, so there is nothing to summarise here.
Identifiers
- Modality record
- text+audio->text+audio
- Catalogue slug
- openai-gpt-audio