# Streaming

> Sending model output incrementally as it is generated instead of waiting for the full response.

- Category: Tools & Protocols
- Stage: Established
- Also called: Streaming response, Token streaming
- Canonical: https://www.joelcsaji.com/ai-dictionary/streaming

## Plain English

The answer starts appearing while the model is still writing it.

## Example

A chat UI displays partial text immediately, then updates as more response events arrive.

## Why it matters

Streaming improves perceived speed for users, but partial output can be harder to moderate or validate.

## Related

- [Latency](https://www.joelcsaji.com/ai-dictionary/latency)
- [Background Mode](https://www.joelcsaji.com/ai-dictionary/background-mode)
- [Tool Result](https://www.joelcsaji.com/ai-dictionary/tool-result)

## Sources

- [OpenAI: Streaming Responses](https://platform.openai.com/docs/guides/streaming-responses)
- [OpenAI: Background Mode](https://developers.openai.com/api/docs/guides/background)