> ## Documentation Index
> Fetch the complete documentation index at: https://docs.scrapellm.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Scrapers

> All ScrapeLLM scrapers, their credit costs, unique features, and supported parameters.

ScrapeLLM supports seven AI providers through a single consistent API. All scrapers use the same authentication, support the same `country` routing, and support async job mode.

## Scraper list and credit costs

| Scraper                                                              | Credits / request | Key unique features                                                  |
| -------------------------------------------------------------------- | ----------------- | -------------------------------------------------------------------- |
| **[ChatGPT](/api-reference/endpoint/monitor-chatgpt)**               | 3                 | `links[]`, `search_queries[]` (query fan-out), `llm_model`           |
| **[Perplexity](/api-reference/endpoint/monitor-perplexity)**         | 3                 | `sources[]` with snippets, `related_questions[]`, `web_search` param |
| **[Grok](/api-reference/endpoint/monitor-grok)**                     | 3                 | `web_sources[]`, `x_results[]` (X/Twitter posts), `mode` param       |
| **[Copilot](/api-reference/endpoint/monitor-copilot)**               | 3                 | `citations[]`, `links[]`, `mode` param (5 modes)                     |
| **[Gemini](/api-reference/endpoint/monitor-gemini)**                 | 3                 | Rich `citations[]` with highlights, favicon, website\_name           |
| **[Google AI Mode](/api-reference/endpoint/monitor-google-ai-mode)** | 3                 | `citations[]`, `search_results[]`, `raw_url`                         |
| **[Amazon Rufus](/api-reference/endpoint/monitor-amazon-rufus)**     | 3                 | `products[]` with ASINs/ratings, `related_questions[]`               |

## Feature comparison

### Query fan-out support

Query fan-out reveals the internal search queries each provider uses to build its answer - invaluable for SEO keyword research.

| Scraper            | Support | Response field              |
| ------------------ | ------- | --------------------------- |
| **ChatGPT**        | ✅ Yes   | `search_queries[]`          |
| **Perplexity**     | ✅ Yes   | Inline in `sources[]`       |
| **Grok**           | ✅ Yes   | Included in `web_sources[]` |
| **Copilot**        | ❌ No    | -                           |
| **Gemini**         | ❌ No    | -                           |
| **Google AI Mode** | ❌ No    | -                           |
| **Amazon Rufus**   | N/A     | Product-based, not text     |

### Source citation support

| Scraper            | Citation field                    | Extra metadata                                                       |
| ------------------ | --------------------------------- | -------------------------------------------------------------------- |
| **ChatGPT**        | `links[]`                         | `url`, `text`                                                        |
| **Perplexity**     | `sources[]`                       | `url`, `title`, `snippet`                                            |
| **Grok**           | `web_sources[]`                   | `url`, `title`, `preview`                                            |
| **Copilot**        | `citations[]`, `links[]`          | `url`, `title`                                                       |
| **Gemini**         | `citations[]`                     | `url`, `title`, `snippet`, `highlights[]`, `favicon`, `website_name` |
| **Google AI Mode** | `citations[]`, `search_results[]` | `url`, `title`, `snippet`, `thumbnail`, `favicon`                    |
| **Amazon Rufus**   | N/A                               | Product-based response                                               |

### Mode parameter support

Some scrapers accept a `mode` parameter to change response behaviour:

| Scraper     | Modes                                                               |
| ----------- | ------------------------------------------------------------------- |
| **Grok**    | `MODEL_MODE_AUTO` (default), `MODEL_MODE_FAST`, `MODEL_MODE_EXPERT` |
| **Copilot** | `search` (default), `smart`, `chat`, `reasoning`, `study`           |

### X (Twitter) integration

Grok is the only scraper with native X integration. The `x_results[]` field returns real X posts that Grok surfaces alongside its answer - a unique signal unavailable from any other provider.

### Amazon Rufus

The Amazon Rufus scraper is fundamentally different from the other AI scrapers - it returns **structured product data** rather than a text answer:

* `products[]` - each with ASIN, title, rating, review count, image URL, category, product URL, and Rufus editorial tag (e.g. "Best overall")
* `related_questions[]` - Rufus-generated follow-up shopping queries
* `interim_messages[]` - Rufus loading-state messages

Supports 11 Amazon marketplaces via the `country` parameter: US, GB, DE, ES, FR, IT, JP, AU, IN, MX, BR.

## Common questions

### Which scraper should I use for brand monitoring?

All text-based scrapers (ChatGPT, Perplexity, Grok, Copilot, Gemini, Google AI Mode) are suitable. For comprehensive AI brand visibility, query the same prompt across multiple scrapers and compare how often and how prominently your brand appears.

### Why use Grok specifically?

Grok's `x_results[]` field exposes real X (Twitter) posts - a signal no other scraper provides. This lets you cross-reference AI recommendations with live social conversation.

### Is Amazon Rufus the same as the Amazon Product Advertising API?

No. ScrapeLLM captures what Amazon Rufus - Amazon's AI shopping assistant - shows to users. This includes editorial positioning (e.g. "Best overall"), curated recommendation sets, and follow-up questions, which are not available through any official Amazon API.

### Do credits roll over?

No. Credits reset at the start of each billing cycle. See [Billing](/guides/billing) for details.