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

# Overview

> Public endpoints, protocol families, and how to choose the right PipeLLM route

PipeLLM exposes native public routes and converter routes. Use this section when
you need the exact path, headers, request format, error behavior, and rate
limit rules for PipeLLM's public API.

## Public Endpoint Families

| Request format          | Native route                             | Converter route                                 | Use this when                                                                                                                                    |
| ----------------------- | ---------------------------------------- | ----------------------------------------------- | ------------------------------------------------------------------------------------------------------------------------------------------------ |
| OpenAI Chat Completions | `/v1/chat/completions`                   | `/openai/v1/chat/completions`                   | Use the native route for OpenAI-compatible platforms. Use the converter route to keep OpenAI format while calling Anthropic or Gemini models.    |
| OpenAI Responses        | `/v1/responses`                          | Not available yet                               | Use the native route for OpenAI-compatible platforms. For cross-protocol calls today, use the OpenAI Chat Completions converter instead.         |
| Anthropic Messages      | `/v1/messages`                           | `/anthropic/v1/messages`                        | Use the native route for Anthropic-compatible platforms. Use the converter route to keep Anthropic format while calling OpenAI or Gemini models. |
| Gemini Generate Content | `/v1beta/models/{model}:generateContent` | `/gemini/v1beta/models/{model}:generateContent` | Use the native route for Gemini platforms. Use the converter route to keep Gemini format while calling OpenAI or Anthropic models.               |

For the detailed routing rules and protocol mismatch behavior, see
[Routing & Protocols](/guides/routing-protocols).

## Authentication

All public routes use your PipeLLM API key with protocol-specific headers:

| API Format        | Header                                   |
| ----------------- | ---------------------------------------- |
| OpenAI Compatible | `Authorization: Bearer $PIPELLM_API_KEY` |
| Anthropic         | `x-api-key: $PIPELLM_API_KEY`            |
| Gemini            | `x-goog-api-key: $PIPELLM_API_KEY`       |

## Base URLs

| Route type           | Base URL                           |
| -------------------- | ---------------------------------- |
| Native public routes | `https://api.pipellm.ai`           |
| OpenAI converter     | `https://api.pipellm.ai/openai/v1` |
| Anthropic converter  | `https://api.pipellm.ai/anthropic` |
| Gemini converter     | `https://api.pipellm.ai/gemini`    |

## Start Here

<Columns cols={2}>
  <Card title="List Models" icon="list" href="/api-reference/list-models">
    Discover the model IDs available to your account
  </Card>

  <Card title="Routing & Protocols" icon="route" href="/guides/routing-protocols">
    Choose the correct free route or converter route
  </Card>

  <Card title="Errors" icon="triangle-exclamation" href="/guides/errors">
    Understand error types, status codes, and retry behavior
  </Card>

  <Card title="Rate Limits" icon="gauge" href="/guides/rate-limits">
    Review request limits, balance protection, and headers
  </Card>
</Columns>

## Endpoint References

<Columns cols={3}>
  <Card title="OpenAI Compatible" icon="bolt" href="/api-reference/openai/overview">
    `/v1/chat/completions` and `/v1/responses`
  </Card>

  <Card title="Anthropic" icon="message" href="/api-reference/anthropic/overview">
    `/v1/messages`
  </Card>

  <Card title="Gemini" icon="sparkles" href="/api-reference/gemini/overview">
    `/v1beta/models/{model}:generateContent`
  </Card>
</Columns>