Skip to content
Source
Sign inSign up
TanStack/ai
Open History to see the last commit

toStreamResponse.md

55 lines1.2 KB

Function: toStreamResponse()

function toStreamResponse(stream, init?): Response;

Defined in: stream-to-response.ts:213

Create a streaming HTTP response from a StreamChunk async iterable Includes proper headers for Server-Sent Events

Parameters

stream

AsyncIterable<StreamChunk>

AsyncIterable of StreamChunks from chat()

init?

ResponseInit & object

Optional Response initialization options

Returns

Response

Response object with SSE headers and streaming body

Deprecated

Use toServerSentEventsStream instead. This function will be removed in a future version.

Example

export async function POST(request: Request) {
  const { messages } = await request.json();
  const abortController = new AbortController();
  const stream = chat({
    adapter: openaiText(),
    model: "gpt-4o",
    messages,
    options: { abortSignal: abortController.signal }
  });
  return toStreamResponse(stream, undefined, abortController);
}