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

# Overview

> Web search plus specialised and proprietary sources through one API, built for AI knowledge work.

Valyu is search built for AI knowledge work. Most search APIs cover the open web; agents working in finance, life sciences, legal, and research need sources general web search doesn't serve well. Valyu gives your agent one API for both - web search alongside deep, full-text retrieval across specialised sources like SEC filings, clinical trials, patents, journals, and market data - and many more.

Agent note: Valyu's full docs index is at [/llms.txt](https://docs.valyu.ai/llms.txt) (full text at /llms-full.txt); append `.md` to any page URL for Markdown; hosted docs MCP at [https://docs.valyu.ai/mcp](https://docs.valyu.ai/mcp); integration reference at [/AGENTS.md](https://docs.valyu.ai/AGENTS.md).

<Note>
  Get **\$10 free credits** (\$20 with a work email) when you sign up at
  [Valyu](https://platform.valyu.ai). Usable across every source. No card required.
</Note>

## What you get

* **Real-time web search** - the latest from across the internet.
* **Specialised & proprietary sources** - academic papers, SEC filings, clinical trials, patents, drug discovery, economic data, and many more.
* **Financial data** - stock prices, fundamentals, and market metrics.
* **Intelligent ranking** - only the most relevant results, ready to feed an LLM.
* **Transparent pricing** - pay per retrieval, no subscription required to start.

## How Valyu fits your stack

One API key, four building blocks, one retrieval layer underneath - your agent reaches the web and proprietary sources through the same interface.

```mermaid theme={null}
flowchart LR
    Agent["Your AI agent"] --> Valyu["Valyu API"]
    Valyu --> Search["Search"]
    Valyu --> Contents["Contents"]
    Valyu --> Answer["Answer"]
    Valyu --> DeepResearch["DeepResearch"]
    DeepResearch --> Workflows["Workflows · Beta"]
    Search --> Sources[("Web + specialised sources<br/>filings · trials · patents · journals · markets · & more")]
    Contents --> Sources
    Answer --> Sources
    DeepResearch --> Sources
```

Start with **Search** for retrieval you control, or hand the whole task to **DeepResearch** and get back a cited report.

## Get started

<CardGroup cols={2}>
  <Card title="Quickstart" icon="bolt" href="/quickstart">
    Get searching in minutes
  </Card>

  <Card title="API Reference" icon="code" href="/api-reference/endpoint/search">
    Explore endpoints and parameters
  </Card>
</CardGroup>

## Integrations

<CardGroup cols={3}>
  <Card title="MCP Server" icon="robot" href="/integrations/mcp-server">
    Remote MCP for Claude, Cursor, and any MCP client
  </Card>

  <Card title="CLI" icon="terminal" href="/integrations/cli">
    Search and research from your terminal
  </Card>

  <Card title="LM Studio" icon="microchip" href="/integrations/lmstudio">
    Web search for local LLMs
  </Card>

  <Card title="Slack" icon="slack" iconType="brands" href="/integrations/slack-bot">
    Scheduled research in Slack
  </Card>

  <Card title="Vercel AI SDK" icon="bolt" href="/integrations/vercel-ai-sdk">
    Tools for the AI SDK
  </Card>

  <Card title="n8n" icon="diagram-project" href="/integrations/n8n">
    Valyu nodes for n8n workflows
  </Card>
</CardGroup>

## Benchmarks

We outperform Google, Parallel, and Exa on SimpleQA and FreshQA, and across domain-specific benchmarks in finance, healthcare, and economics.

<Frame caption="FreshQA: Valyu vs Google, Parallel, and Exa on 600 time-sensitive queries.">
  ![FreshQA benchmark results comparing Valyu against Google, Parallel, and Exa](https://4ealzrotsszllxtz.public.blob.vercel-storage.com/freshqa_valyu-1400x871.png)
</Frame>

| **Benchmark**                              | **Valyu** | **Parallel** | **Exa** | **Google** |                 |
| ------------------------------------------ | --------- | ------------ | ------- | ---------- | --------------- |
| **FreshQA** (600 time-sensitive queries)   | 79 %      | 52 %         | 24 %    | 39 %       | ([valyu.ai][1]) |
| **SimpleQA** (4,326 factual questions)     | 94 %      | 93 %         | 91 %    | 38 %       | ([valyu.ai][1]) |
| **Finance** (120 finance questions)        | 73 %      | 67 %         | 63 %    | 55 %       | ([valyu.ai][1]) |
| **Economics** (100 economics questions)    | 73 %      | 52 %         | 45 %    | 43 %       | ([valyu.ai][1]) |
| **MedAgent** (562 complex medical queries) | 48 %      | 42 %         | 44 %    | 45 %       | ([valyu.ai][1]) |

[1]: https://www.valyu.ai/blogs/benchmarking-search-apis-for-ai-agents?utm_source=chatgpt.com "Benchmarking Search APIs for AI Agents | Valyu | Valyu"

Read the full breakdown in our  [benchmarking blog post](https://www.valyu.ai/blogs/benchmarking-search-apis-for-ai-agents).