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

# Parallel

> AI-native deep research and search API for finding, extracting, and analyzing information across the web.

AI-native deep research and search API for finding, extracting, and analyzing information across the web. Use when a workflow needs verified factual answers, structured data extraction from URLs, or automated research tasks that go beyond simple search. Tasks and findall are async with polling.

4 endpoints available through Lava's AI Gateway. See the [Parallel API docs](https://docs.parallel.ai) for full documentation.

<Info>Supports both **managed** (Lava's API keys) and **unmanaged** (bring your own credentials) mode.</Info>

## Endpoints

### Search the web with AI-powered relevance

**POST** `https://api.parallel.ai/v1beta/search` — \$0.005 / request

<Tabs>
  <Tab title="SDK">
    ```typescript theme={null}
    const data = await lava.gateway('https://api.parallel.ai/v1beta/search', {
      body: {
    "objective": "Find latest AI research from 2026",
    "search_queries": [
      "latest AI research papers 2026",
      "AI breakthroughs 2026"
    ]
    },
    });
    ```
  </Tab>

  <Tab title="cURL">
    ```bash theme={null}
    curl -X POST "https://api.lava.so/v1/forward?u=https%3A%2F%2Fapi.parallel.ai%2Fv1beta%2Fsearch" \
      -H "Authorization: Bearer $LAVA_SECRET_KEY" \
      -H "Content-Type: application/json" \
      -d '{"objective":"Find latest AI research from 2026","search_queries":["latest AI research papers 2026","AI breakthroughs 2026"]}'
    ```
  </Tab>
</Tabs>

### Extract structured data from a URL

**POST** `https://api.parallel.ai/v1beta/extract` — \$0.001 / request

<Tabs>
  <Tab title="SDK">
    ```typescript theme={null}
    const data = await lava.gateway('https://api.parallel.ai/v1beta/extract', {
      body: {
    "urls": [
      "https://example.com"
    ],
    "objective": "Summarize the page content"
    },
    });
    ```
  </Tab>

  <Tab title="cURL">
    ```bash theme={null}
    curl -X POST "https://api.lava.so/v1/forward?u=https%3A%2F%2Fapi.parallel.ai%2Fv1beta%2Fextract" \
      -H "Authorization: Bearer $LAVA_SECRET_KEY" \
      -H "Content-Type: application/json" \
      -d '{"urls":["https://example.com"],"objective":"Summarize the page content"}'
    ```
  </Tab>
</Tabs>

### Submit a deep research task (async, poll for results)

**POST** `https://api.parallel.ai/v1/tasks/runs` — \$0.025 / request

<Tabs>
  <Tab title="SDK">
    ```typescript theme={null}
    const data = await lava.gateway('https://api.parallel.ai/v1/tasks/runs', {
      body: {
    "input": "Find all Series A funding rounds in Q1 2026",
    "processor": "base"
    },
    });
    ```
  </Tab>

  <Tab title="cURL">
    ```bash theme={null}
    curl -X POST "https://api.lava.so/v1/forward?u=https%3A%2F%2Fapi.parallel.ai%2Fv1%2Ftasks%2Fruns" \
      -H "Authorization: Bearer $LAVA_SECRET_KEY" \
      -H "Content-Type: application/json" \
      -d '{"input":"Find all Series A funding rounds in Q1 2026","processor":"base"}'
    ```
  </Tab>
</Tabs>

### Find all instances matching structured match conditions (async, poll for results). Requires the `parallel-beta` header — injected by Parallel's custom auth handler.

**POST** `https://api.parallel.ai/v1beta/findall/runs` — \$0.03 / request

<Tabs>
  <Tab title="SDK">
    ```typescript theme={null}
    const data = await lava.gateway('https://api.parallel.ai/v1beta/findall/runs', {
      body: {
    "objective": "Find AI startups founded in 2025",
    "entity_type": "companies",
    "match_conditions": [
      {
        "name": "founded_2025",
        "description": "Founded in 2025"
      }
    ],
    "generator": "base",
    "match_limit": 5
    },
    });
    ```
  </Tab>

  <Tab title="cURL">
    ```bash theme={null}
    curl -X POST "https://api.lava.so/v1/forward?u=https%3A%2F%2Fapi.parallel.ai%2Fv1beta%2Ffindall%2Fruns" \
      -H "Authorization: Bearer $LAVA_SECRET_KEY" \
      -H "Content-Type: application/json" \
      -d '{"objective":"Find AI startups founded in 2025","entity_type":"companies","match_conditions":[{"name":"founded_2025","description":"Founded in 2025"}],"generator":"base","match_limit":5}'
    ```
  </Tab>
</Tabs>

## Next Steps

<CardGroup cols={2}>
  <Card title="All Providers" icon="grid" href="/gateway/supported-providers">
    Browse all supported AI providers
  </Card>

  <Card title="Forward Proxy" icon="route" href="/gateway/forward-proxy">
    Learn how to construct proxy URLs and authenticate requests
  </Card>
</CardGroup>