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

# Web Search Premier

> Higher-accuracy web search, powered by deeper retrieval and reranking.

For AI agents: [docs.octen.ai/capabilities/web-search-premier.md](https://docs.octen.ai/capabilities/web-search-premier.md)

Some questions need the exact page that holds the answer: a specific fact, a niche document, or a page an agent was asked to find.

Web Search Premier searches more broadly and ranks results more carefully, so the page with the answer ranks higher.

For the full list of parameters, see the [Web Search Premier API reference](https://docs.octen.ai/api-reference/search-premier.md).

## Why Web Search Premier

* **Answer pages rank higher.** The page that holds the answer is more likely to appear in the top results.
* **Highlights go straight to the answer.** Highlights pick the passage most relevant to the query from the whole page, so an answer deep inside a page still comes back directly.
* **Broader coverage.** Each query reaches more sources, including niche and long-tail pages.
* **Drop-in for Web Search.** Same parameters, same response. Change the endpoint and you are done.

For low-latency lookups, use [Web Search](https://docs.octen.ai/capabilities/web-search.md) instead.

## How It Works

1. Send a query, using the same body as Web Search.
2. Octen searches its live index more broadly and ranks results more carefully.
3. The response returns ranked results, each with a query-relevant highlight and optional full page content.

## Scenarios

### Answer an agent's question

Pass the question as is. Results come back with highlights ready for the model.

```bash theme={null}
curl -X POST https://api.octen.ai/search-premier \
  -H "Content-Type: application/json" \
  -H "x-api-key: YOUR_API_KEY" \
  -d '{
    "query": "How long does Kubernetes wait after SIGTERM before it force kills a pod?",
    "count": 5
  }'
```

### Look up a specific fact

Ask for a detail that only appears deep inside a page. Raise `highlight.max_tokens` to keep more of the surrounding text.

```json theme={null}
{
  "query": "Kafka default retention period for topics",
  "count": 5,
  "highlight": { "enable": true, "max_tokens": 1024 }
}
```

### Find a named page

Search for a report, filing, or document by its name.

```json theme={null}
{
  "query": "Basel III endgame proposal Federal Reserve 2023 press release",
  "count": 5
}
```

### Get the original text

Turn on `full_content` to read the whole page. The full content of the first 10 results per search is free; only additional results are billed.

```json theme={null}
{
  "query": "Apple revenue for its last full fiscal year",
  "count": 5,
  "full_content": { "enable": true, "max_tokens": 4096 }
}
```

## Next Steps

> [card] [Try in Console](https://octen.ai/platform/web-search-premier)
>
> Run Web Search Premier live in the Octen console.

> [card] [Web Search Premier API Reference](https://docs.octen.ai/api-reference/search-premier.md)
>
> Full request/response schema.

***

Last modified on November 20, 2026


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.