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

# Introduction

> HeyBee helps you decide which AI output is better, backed by real human judgments.

You have five prompt variants, three checkpoints, or two generation settings, and you need to know which one wins. HeyBee collects side-by-side human comparisons of your outputs and turns them into a clear answer: what wins, by how much, and with what confidence.

It works with text, images, audio, and video. Share a voting link with anyone, watch votes arrive, and get a result that tells you what to do next.

<CardGroup cols={2}>
  <Card title="Quickstart" icon="rocket" href="/product/getting-started">
    Go from a question to your first evidence-backed decision.
  </Card>

  <Card title="Create an evaluation" icon="flask" href="/product/evaluations">
    Define what you are comparing and how voters should judge it.
  </Card>

  <Card title="Collect votes" icon="users" href="/product/voting">
    Share voting links and see exactly what voters see.
  </Card>

  <Card title="Read your results" icon="chart-line" href="/product/results">
    Understand the verdict, the confidence score, and the next action.
  </Card>
</CardGroup>

## Build with HeyBee

Run the whole lifecycle from code: create evaluations, upload outputs, open collection, and export results.

<CardGroup cols={3}>
  <Card title="Python SDK" icon="python" href="/developers/python-sdk">
    Typed sync and async clients.
  </Card>

  <Card title="CLI" icon="terminal" href="/developers/cli">
    Every workflow from your shell, with stable JSON output.
  </Card>

  <Card title="Agent Skills" icon="robot" href="/developers/skills">
    Let coding agents run evaluations for you.
  </Card>
</CardGroup>
