Integrations

Most integrations need the same three values:

text

Base URL: https://api.rout.my/v1
API key:  sk_your_key_here
Model:    openai/gpt-5.4

That base URL is for OpenAI-compatible clients. Anthropic-native clients such as Claude Code and the Anthropic SDK usually want https://api.rout.my as the configured base URL because they append /v1/messages themselves.

Use these defaults unless you intentionally need another model:

Client type	Base URL	Model
OpenAI-compatible clients	`https://api.rout.my/v1`	`openai/gpt-5.4`
Codex	`https://api.rout.my/v1`	`openai/gpt-5.4`
Claude Code / Anthropic SDK	`https://api.rout.my`	`anthropic/claude-sonnet-4.6`

Use GET /v1/models with your key only when you want to replace the default with another exact model ID.

OpenAI SDK

Python and Node examples with custom base URL.

SillyTavern

Custom OpenAI-compatible Chat Completion source.

Janitor AI

Proxy setup with the full /v1/chat/completions URL.

OpenClaw

openai-completions provider config with rout.my as baseUrl.

Hermes Agent

Custom provider setup with prompt caching enabled.

Codex / CLI tools

Custom Codex provider config with Responses wire API.

Claude Code / Anthropic SDK

Anthropic Messages setup for Claude Code, Python, and TypeScript.

Quick Check

bash

curl https://api.rout.my/v1/models \
  -H "Authorization: Bearer $ROUTMY_API_KEY"

If the app can set a base URL and send a bearer token, it can usually talk to rout.my.

Integrations ​