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

# Introduction

> URL-level classification against the ContentGraph index. Priced per URL, never per token.

Classify tells you what a page is about, not what site it lives on. The API classifies any URL against ContentGraph: an embedding-native semantic index of the open web, labeled at the page level with the IAB 3.1 taxonomy.

## Two ways to classify

<CardGroup cols={2}>
  <Card title="Lookup" icon="bolt">
    The URL is already in ContentGraph. Classifications return inline in under 50ms, billed at the lookup rate.
  </Card>

  <Card title="Live Classify" icon="globe">
    The URL is new to us. We fetch it, embed it, and classify it in seconds. Poll the job for results. Every live result joins the index.
  </Card>
</CardGroup>

The response tells you which path served each URL, and billing follows the path.

## Endpoints

Three endpoints, one auth scheme:

| Endpoint                | What it does                         |
| ----------------------- | ------------------------------------ |
| `POST /v1/classify`     | Classify up to 1000 URLs per request |
| `GET /v1/jobs/{job_id}` | Poll a live-classification job       |
| `GET /health`           | Unauthenticated liveness check       |

Try them in the [API Reference](/api-reference) with your key.