IStreamingGenerator<T>

Interfaces · AiDotNet.Interfaces

A generator that can stream its output incrementally (token/chunk deltas) as it is produced, in addition to the buffered Generate.

Implements: IGenerator<T>

Known implementations: ChatClientGenerator<T>

For Beginners

instead of waiting for the whole answer, you get it piece by piece — like watching text appear as an assistant “types” — by iterating the returned async stream.

How It Works

Streaming lets a RAG answer render as it is generated rather than after the whole response is complete, which is the standard low-latency UX for chat/RAG systems.

Methods

Method Summary
IAsyncEnumerable<string> GenerateStreamAsync(string prompt, CancellationToken cancellationToken) Generates a response to prompt, yielding incremental text deltas as they arrive.