> For the complete documentation index, see [llms.txt](https://docs.umbraco.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.umbraco.com/ai-in-umbraco/19.latest/using-the-api/chat/streaming.md).

# Streaming

Stream chat responses in real-time for a better user experience.

Streaming returns response chunks as they're generated, rather than waiting for the complete response. Streaming provides a better user experience for longer responses.

## Why Use Streaming

| Approach      | Best For                                             |
| ------------- | ---------------------------------------------------- |
| Non-streaming | Short responses, background processing               |
| Streaming     | User-facing chat, long responses, real-time feedback |

Streaming lets users see the response as it's being generated, reducing perceived latency.

## Basic Streaming

{% code title="BasicStreaming.cs" %}

```csharp
using Microsoft.Extensions.AI;
using Umbraco.AI.Core.Chat;

public class StreamingExample
{
    private readonly IAIChatService _chatService;

    public StreamingExample(IAIChatService chatService)
    {
        _chatService = chatService;
    }

    public async Task StreamToConsole(string question)
    {
        var messages = new List<ChatMessage>
        {
            new(ChatRole.User, question)
        };

        await foreach (var update in _chatService.StreamChatResponseAsync(
            chat => chat.WithAlias("console-stream"),
            messages))
        {
            // Each update contains a chunk of text
            Console.Write(update.Text);
        }

        Console.WriteLine(); // New line at end
    }
}
```

{% endcode %}

## Understanding ChatResponseUpdate

Each streamed chunk is a `ChatResponseUpdate`:

| Property       | Type                | Description                            |
| -------------- | ------------------- | -------------------------------------- |
| `Text`         | `string?`           | The text content of this chunk         |
| `Role`         | `ChatRole?`         | The role (usually only in first chunk) |
| `FinishReason` | `ChatFinishReason?` | Set in the final chunk                 |

{% code title="UpdateDetails.cs" %}

```csharp
await foreach (var update in _chatService.StreamChatResponseAsync(
    chat => chat.WithAlias("update-details"),
    messages))
{
    if (update.Text is not null)
    {
        // Append text chunk
        responseBuilder.Append(update.Text);
    }

    if (update.FinishReason is not null)
    {
        // This is the last chunk
        Console.WriteLine($"Finished: {update.FinishReason}");
    }
}
```

{% endcode %}

## Streaming in an API Controller

Return a streaming response to the client:

{% code title="StreamingController.cs" %}

```csharp
[ApiController]
[Route("api/chat")]
public class ChatController : UmbracoApiController
{
    private readonly IAIChatService _chatService;

    public ChatController(IAIChatService chatService)
    {
        _chatService = chatService;
    }

    [HttpPost("stream")]
    public async Task StreamChat([FromBody] ChatRequest request)
    {
        Response.Headers.Append("Content-Type", "text/event-stream");
        Response.Headers.Append("Cache-Control", "no-cache");
        Response.Headers.Append("Connection", "keep-alive");

        var messages = new List<ChatMessage>
        {
            new(ChatRole.User, request.Message)
        };

        await foreach (var update in _chatService.StreamChatResponseAsync(
            chat => chat.WithAlias("chat-stream"),
            messages))
        {
            if (update.Text is not null)
            {
                await Response.WriteAsync($"data: {update.Text}\n\n");
                await Response.Body.FlushAsync();
            }
        }

        await Response.WriteAsync("data: [DONE]\n\n");
    }
}

public record ChatRequest(string Message);
```

{% endcode %}

## Collecting the Full Response

To collect the full text while streaming, append each `update.Text` to a `StringBuilder` inside the `await foreach` loop.

{% hint style="info" %}
`StreamChatResponseAsync` accepts a `CancellationToken` parameter to support user cancellation.
{% endhint %}

## Using a Specific Profile

{% code title="StreamWithProfile.cs" %}

```csharp
// With profile ID
await foreach (var update in _chatService.StreamChatResponseAsync(
    chat => chat
        .WithAlias("profiled-stream")
        .WithProfile(profileId),
    messages))
{
    Console.Write(update.Text);
}

// With options
var options = new ChatOptions { Temperature = 0.8f };

await foreach (var update in _chatService.StreamChatResponseAsync(
    chat => chat
        .WithAlias("custom-options-stream")
        .WithChatOptions(options),
    messages))
{
    Console.Write(update.Text);
}
```

{% endcode %}

## Next Steps

{% content-ref url="/pages/xyrLrHgpZz62viguQe5S" %}
[System Prompts](/ai-in-umbraco/19.latest/using-the-api/chat/system-prompts.md)
{% endcontent-ref %}

{% content-ref url="/pages/94HtZGVrX72QYCCEOWqQ" %}
[Advanced Options](/ai-in-umbraco/19.latest/using-the-api/chat/advanced-options.md)
{% endcontent-ref %}


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation by asking a question.

Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters:

```
GET https://docs.umbraco.com/ai-in-umbraco/19.latest/using-the-api/chat/streaming.md?ask=<question>&goal=<user_goal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.