> For the complete documentation index, see [llms.txt](https://help.netnut.io/netnut-documentation/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://help.netnut.io/netnut-documentation/netnut-scraper-apis/llm-scraper/async-requests.md).

# Async Requests

**By default, LLM Scraper requests are processed synchronously.**

In synchronous mode, the customer sends a request and waits for the final model response in the same HTTP request.

For longer-running requests, larger prompts, or production workflows where the customer does not want to keep the HTTP connection open, NetNut’s LLM Scraper supports Async mode.

Async mode is recommended for workflows that require better stability, longer processing time, backend automation, or request tracking.

**Async mode is supported with all LLM Models**

***

### Parameter: `async`

| Field       | Value                                                                                                                                             |
| ----------- | ------------------------------------------------------------------------------------------------------------------------------------------------- |
| Type        | Boolean                                                                                                                                           |
| Default     | `false`                                                                                                                                           |
| Description | When set to `true`, the request is processed asynchronously and the API returns a `response_url` instead of waiting for the final model response. |

***

## How Async Works

When `async=true`, the request flow changes from a direct response flow to a two-step flow.

### Step 1: Send the Async Request

The customer sends a regular LLM Scraper request with `async=true`.

```json
{  
  "engine": "perplexity",
  "prompt": "What are the latest trends in the coffee industry?",
  "async": true
}
```

### Step 2: Receive the Initial Response

The API immediately returns a confirmation response.

```json
{  
    "message": "request received",
    "traceId": "123e4567-e89b-12d3-a456-426614174000",
    "response_url": "https://llm-scraper.netnut.io/search/response/123e4567-e89b-12d3-a456-426614174000"
}
```

### Step 3: Retrieve the Final Response

The customer sends a `GET` request to the returned `response_url`.

```http
GET https://llm-scraper.netnut.io/search/response/123e4567-e89b-12d3-a456-426614174000
```

{% hint style="info" %}
Each response is protected with Basic Auth. Include valid Basic Auth credentials when retrieving it.
{% endhint %}

Once the request is completed, the API returns the final LLM Scraper response.

***

## Response URL Behavior

The `response_url` is used to retrieve the completed async response.

If the request is still processing, the API returns:

```json
{  
  "status": "processing",
  "message": "Request is still processing",
  "traceId": "123e4567-e89b-12d3-a456-426614174000"
}
```

The customer should retry the same `response_url` until the final response is available.

Completed async responses are available through the `response_url` for **2 hours after the request is completed**.

***

## Response Storage

Completed async responses are stored on NetNut internal servers for up to **2 hours after the request is completed**.

During this period, customers can retrieve the final response using the provided `response_url`.

After the storage period expires, the `response_url` may no longer return the completed response.

***

## Async with `callback_url`

Async mode can also be used together with `callback_url`.

When both `async=true` and `callback_url` are provided, the API still returns the initial async response with the `response_url`.

Once processing is completed, NetNut sends a notification to the customer’s `callback_url`.

In this flow, the callback does not include the full final response. Instead, it notifies the customer that the request is complete and includes the `response_url`.

### Example Request

```json
{  
   "model": "perplexity",
   "prompt": "What are the latest trends in the coffee industry?",  
   "async": true,  
   "callback_url": "https://customer-api.com/llm-callback"
}
```

### Example Callback Payload

```json
{
  "traceId": "123e4567-e89b-12d3-a456-426614174000",
  "status_code": 200,
  "response_url": "https://llm-scraper.netnut.io/search/response/123e4567-e89b-12d3-a456-426614174000"
}
```

The customer can then use the `response_url` to retrieve the completed result.

***

## Async vs Callback URL

| Mode                 | Final Response Flow                                                                                         |
| -------------------- | ----------------------------------------------------------------------------------------------------------- |
| Sync                 | Final response is returned in the same request                                                              |
| Async                | Customer retrieves the final response from `response_url`                                                   |
| Callback URL         | Final response is sent directly to the customer’s `callback_url`                                            |
| Async + Callback URL | Callback sends a completion notification with `response_url`; customer retrieves the final response from it |


---

# 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 dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://help.netnut.io/netnut-documentation/netnut-scraper-apis/llm-scraper/async-requests.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
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.