# GPT-Realtime-2

GPT-Realtime-2 is a model from OpenAI, released on 7 May 2026. The context window is 128,000 tokens.

- Developer: OpenAI
- Id: gpt-realtime-2
- Released: 2026-05-07T00:00:00.000Z
- Sighting: 2026-10-07T16:22:11.205Z
  - [evidence link](https://github.com/truefoundry/models/blob/main/providers/microsoft-foundry/gpt-realtime-2.yaml)
- Status: Released
- API availability: 2026-05-07
  - [evidence link](https://developers.openai.com/api/docs/changelog)
  - [evidence link](https://developers.openai.com/api/docs/changelog)
- Canonical: https://alexgetman.com/models/openai/gpt-realtime-2/

## Questions

### What is the context window of GPT-Realtime-2?

The context window is 128,000 tokens.

### When was GPT-Realtime-2 released?

On 7 May 2026.

## Evidence links

- [review](https://developers.openai.com/api/docs/changelog)
- [review](https://developers.openai.com/api/docs/models/gpt-realtime-2)
- [openai-pricing](https://developers.openai.com/api/docs/pricing)
- [truefoundry-azure](https://github.com/truefoundry/models/blob/main/providers/microsoft-foundry/gpt-realtime-2.yaml)
- [Azure Foundry · lifecycle](https://learn.microsoft.com/en-us/azure/foundry/concepts/model-lifecycle-retirement?view=azureml-api-2)
- [models.dev · catalogue](https://models.dev/#gpt-realtime-2)

## Signals

- [GPT-Realtime-2: Catalogue listing](https://alexgetman.com/signals/2026/09/17/gpt-realtime-2/)