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

# Linkup Flash

> _Released: September 2026_

We're releasing **Linkup Flash**, a new `depth` value for the [**Search**](/pages/documentation/endpoints/search/overview) endpoint. `"flash"` is our lowest-latency search mode: ranked sources and snippets in under 200 ms.

### What's new

`"flash"` uses a more direct search pipeline than `"fast"`, `"standard"`, or `"deep"`:

* **`"flash"`**: lowest latency. One piece of information, one direct answer. No query reinterpretation, no scraping.
* **`"fast"`**: higher-quality one-shot retrieval in about a second. Still no agentic planning.
* **`"standard"`**: single-iteration agentic search for queries that span several topics or sources.
* **`"deep"`**: multi-iteration agentic search-and-scrape.

### How to use

Set `depth` to `"flash"` in your API request:

<CodeGroup>
  ```shell curl theme={"system"}
  curl --request POST \
    --url https://api.linkup.so/v1/search \
    --header 'Authorization: Bearer {{LINKUP_API_KEY}}' \
    --header 'Content-Type: application/json' \
    --data '{
    "q": "What changed in the latest EU AI Act guidance?",
    "depth": "flash",
    "outputType": "searchResults"
  }'
  ```

  ```python python theme={"system"}
  from linkup import LinkupClient

  client = LinkupClient(api_key="{{LINKUP_API_KEY}}")

  response = client.search(
      query="What changed in the latest EU AI Act guidance?",
      depth="flash",
      output_type="searchResults",
  )
  print(response)
  ```

  ```javascript js theme={"system"}
  import { LinkupClient } from 'linkup-sdk';

  const client = new LinkupClient({
    apiKey: '{{LINKUP_API_KEY}}',
  });

  const response = await client.search({
    query: "What changed in the latest EU AI Act guidance?",
    depth: "flash",
    outputType: "searchResults",
  });
  console.log(response);
  ```
</CodeGroup>

<Tip>
  `"flash"` and `"fast"` pass the query to the index as-is. Keep prompts short and keyword-shaped. For instruction-style prompts (scrape this URL, then search again), use `"standard"` or `"deep"`.
</Tip>

### When to use each depth

| Depth           | Best for                                                          | Latency  |
| --------------- | ----------------------------------------------------------------- | -------- |
| `"flash"`       | Simple, focused queries: one piece of information, lowest latency | \<200 ms |
| `"fast"` (beta) | Higher-quality one-shot lookups when \~1s is acceptable           | \~1s     |
| `"standard"`    | Instruction-style retrieval, several topics or one provided URL   | 1–3s     |
| `"deep"`        | Sequential search-then-scrape chains                              | 5–30s    |

### Pricing

`"flash"` costs \*\*$0.005 per call** with `outputType: "searchResults"`, the same as `"fast"` and `"standard"`. `"sourcedAnswer"` and `"structured"` are $0.006 per call.

For more details, see [Search overview](/pages/documentation/endpoints/search/overview).