Whisper
Generally availableProprietary448K context
Summary
Whisper is a proprietary language model from OpenAI, released on 1 Mar 2023. It accepts audio and generates text, with a 448K-token context window and up to 4K output tokens per response.
General-purpose speech recognition model for transcription and translation
Last verified 21 Oct 2026. Every figure on this page links to its source.
Specifications
| Developer | OpenAI |
|---|---|
| Release date | 1 Mar 2023 [models.dev] |
| Status | Generally available [models.dev] |
| Weights | Proprietary (API only) [models.dev] |
| Context window | 448K (448,000 tokens) [models.dev] |
| Max output | 4K tokens [models.dev] |
| Input | Audio [models.dev] |
| Output | Text [models.dev] |
| Reasoning mode | No [models.dev] |
| Tool calling | No [models.dev] |
| Structured output | No [models.dev] |
Whisper API pricing by provider
USD per million tokens. Sorted by input price.
| Provider | Input | Output | Cached input | Context | Source |
|---|---|---|---|---|---|
| Vercel AI Gateway openai/whisper-1 | – | – | – | – | models.dev, 21 Oct 2026 |
Frequently asked questions
What is the context window of Whisper?
Whisper has a context window of 448,000 tokens (448K) and can generate up to 4,096 tokens in a single response, according to models.dev.
Is Whisper open source?
No. Whisper is a proprietary model; OpenAI has not released its weights. It is available through APIs including Vercel AI Gateway.
When was Whisper released?
OpenAI released Whisper on 1 Mar 2023, according to models.dev.
Change history
- Added to the directory
- Now listed on Vercel AI Gateway