AI Applications

Connect AI Router to popular AI apps and tools using one API key and one base URL.

Overview

AI Router is an OpenAI-compatible AI gateway: you get a single API key and a single base URL that route requests to many model providers. Because the API surface follows the OpenAI convention, most third-party AI apps and tools can talk to AI Router with no special integration — you simply point them at your AI Router endpoint and paste in a token.

This page lists applications with dedicated step-by-step guides. Any client that supports a custom OpenAI-compatible base URL should also work even if it isn't listed here.

Supported Applications

AionUi

Free, local, open-source 24/7 Cowork and OpenClaw for Gemini CLI, Claude Code, Codex, OpenCode, Qwen Code, Goose CLI, Auggie, and more.

CC Switch

One-click fill from the AI Router token page; configure Claude, Codex, or Gemini with main and variant models in the popup.

Cherry Studio

A powerful AI assistant client that supports multi-model conversations.

DeepChat

An open-source, cross-platform multi-model AI client with one-click AI Router provider import via DeepLink.

Memoh

A containerized AI agent platform where each bot runs in its own container with long-term memory and multi-channel support.

OpenClaw

A self-hosted AI assistant platform — install OpenClaw, connect to AI Router, and manage agents across Feishu, Discord, Slack, and more.

Fluent Read

An AI-powered smart reading and translation assistant.

LangBot

A large language model-based chatbot framework.

Luna Translator

A real-time translation tool for games and documents.

AstrBot

An open-source, all-in-one Agent chatbot platform.

Claude Code

An Anthropic Claude-powered coding assistant integration.

Codex CLI

A command-line AI coding assistant.

Factory Droid CLI

An AI agent tool for automating workflows.

Integration Guide

Most applications connect to AI Router with three pieces of information:

API base URL — your AI Router service address. The OpenAI-compatible endpoint is:

https://ai.sammier.com/v1

Some apps expect the full path ending in /v1; others expect only the host (https://ai.sammier.com) and append the path themselves. If one form fails, try the other.

API key — a token you generate on the AI Router console's token page. It is sent as a standard bearer credential, for example Authorization: Bearer <your-token>.

Model name — the model (or model group) you want to call. Use the exact model identifier exposed by your AI Router instance.

A quick way to confirm connectivity is a plain OpenAI-style request:

curl https://ai.sammier.com/v1/chat/completions \
  -H "Authorization: Bearer <your-token>" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "<your-model-name>",
    "messages": [{"role": "user", "content": "Hello"}]
  }'

You can also list the models available to your token:

curl https://ai.sammier.com/v1/models \
  -H "Authorization: Bearer <your-token>"

For app-specific steps (where to paste each field, DeepLink import, etc.), open the relevant guide above.

The base URL above is a placeholder for documentation. Replace it with your actual AI Router deployment address. Available models, model groups, and any usage limits depend on your instance configuration (details TBD / 待补).

Don't see your app, or a guide is out of date? Open an issue on the source repository: github.com/ssgzy/Router-new-api.