XEDOC API Documentation | Setup, Models and API Reference

Make the connection

#

Your first request

Create a key and make a working API call.

Set up your tools

Let the XEDOC assistant connect your preferred clients.

Explore the API

Familiar request formats, with one XEDOC endpoint.

One endpoint. Your choice of client.

#

XEDOC connects supported models to Codex, Claude Code, OpenCode, and applications built with OpenAI- or Anthropic-compatible clients. Your application uses a XEDOC key; provider credentials stay on the server.

BASE URL

OpenAI-compatible clientshttps://api.xedoc.top/v1

BASE URL

Anthropic-compatible clientshttps://api.xedoc.top

Use the right base URL

OpenAI-compatible clients need /v1. Anthropic clients add /v1/messages themselves, so their base URL ends at the hostname.

Meet your tools where they are

#

Codex CLI

Connect from your terminal.

Codex App

Use XEDOC in the desktop app.

Claude Code

Configure the Messages API.

Claude Desktop

Create a third-party inference connection.

OpenCode

Add a provider and discover the current models.

Your application

Start with cURL, Python or JavaScript.

Before you begin

#

Create an account and make an API key. API access is enabled automatically for password, GitHub and Discord registrations. Accounts start at 0; add token credit in Overview before generating. Administrators can disable access for individual accounts. A valid key can list models before the account is funded.

Your account, your key

Use a key from Dashboard → API keys. A password, OAuth token or upstream provider key is not your XEDOC API key.

Keep building

Your first XEDOC API request

Create an API key, check your model access and send a Responses API request with cURL or PowerShell.

Set up your AI coding tools

Connect Codex, Claude Code, Claude Desktop and OpenCode with the XEDOC setup assistant on Windows, macOS or Linux.