OpenAI

gpt-realtime-whisper

Generally availableProprietary16K context

Summary

gpt-realtime-whisper is a proprietary language model from OpenAI, released on 7 May 2026. It accepts audio and generates text, with a 16K-token context window and up to 2K output tokens per response.

Streaming speech-to-text model for low-latency transcript deltas from live audio

Last verified 18 Nov 2026. Every figure on this page links to its source.

Specifications

DeveloperOpenAI
Release date7 May 2026 [models.dev]
StatusGenerally available [models.dev]
WeightsProprietary (API only) [models.dev]
Context window16K (16,000 tokens) [models.dev]
Max output2K tokens [models.dev]
InputAudio [models.dev]
OutputText [models.dev]
Reasoning modeNo [models.dev]
Tool callingNo [models.dev]

gpt-realtime-whisper API pricing by provider

USD per million tokens. Sorted by input price.

ProviderInputOutputCached inputContextSource
Vercel AI Gateway
openai/gpt-realtime-whisper
–––16Kmodels.dev, 18 Nov 2026

Frequently asked questions

What is the context window of gpt-realtime-whisper?

gpt-realtime-whisper has a context window of 16,000 tokens (16K) and can generate up to 2,000 tokens in a single response, according to models.dev.

Is gpt-realtime-whisper open source?

No. gpt-realtime-whisper is a proprietary model; OpenAI has not released its weights. It is available through APIs including Vercel AI Gateway.

When was gpt-realtime-whisper released?

OpenAI released gpt-realtime-whisper on 7 May 2026, according to models.dev.

Similar models from other labs

Comparable on output type, weights, release date, price and context window. Prices are the lowest listed, USD per million tokens.

ModelLabReleasedContextInput $/1MOutput $/1MWeights
Gemini 3.5 Live Translate PreviewGoogle DeepMind9 Jun 202616K$3.50$21Closed
Grok Voice STT 1.0xAI4 Aug 202615K––Closed
Perceptron Mk1Perceptron12 May 202633K$0.15$1.50Closed
Gemini Embedding 2Google DeepMind22 Apr 20268K$0.20–Closed
Grok Voice STT 1.0xAI17 Sept 202615K––Closed
Mercury Edit 2Inception Labs30 Mar 202632K$0.25$0.75Closed

Change history

  • Context window changed from none to 16K
  • Max output changed from none to 2K

Sources