Models Hub
Agent Tools

Connecting Agent Tools

Integration guides for AI agents and client tools that already support connecting to Models Hub.

Edit this page

Models Hub is compatible with many popular AI applications and tools. Below is a list of verified supported applications along with their configuration guides.

Supported Applications

One-click configuration for mainstream agent platforms, so you can easily use a variety of agent tools

Use the Models Hub CLI to automatically configure Claude Code, Codex, Gemini CLI, and OpenClaw.

AstrBot - Agent Chatbot

A multi-platform chatbot and agent framework. Add an OpenAI provider pointed at Models Hub to use it on QQ, Feishu, and more.

CC Switch

One-click import from the Models Hub token page, writing straight into Claude Code, Codex, or Gemini CLI.

Chatbox

Cross-platform desktop AI chat client. Add an OpenAI API-compatible provider to connect to Models Hub.

Cherry Studio

Desktop AI assistant client. Add Models Hub as a provider and switch between Claude and GPT with one key.

Claude Code

Anthropic's terminal coding assistant. Connect to Models Hub with one command or four env vars, with verification and troubleshooting.

Cline

VS Code coding agent. Connect via OpenAI Compatible or Anthropic with a custom base URL.

CodeBuddy

Tencent Cloud's terminal coding assistant, modeled on Claude Code — connect via env vars or settings.json.

Codex CLI

OpenAI's terminal coding agent. Connect to Models Hub with one click or by editing config.toml.

Gemini CLI

Google Gemini CLI, which can have its API Key, default model, and base URL written in with one click via the Models Hub CLI.

Copilot CLI

GitHub's terminal assistant. BYOK env vars, with the anthropic protocol type recommended to avoid reasoning errors.

Crush

An open-source terminal AI coding agent by Charm — declare a custom provider in its crushrc script.

Cursor

AI-native code editor. Connect via Override OpenAI Base URL and use Claude and GPT with one key.

Factory Droid CLI

A terminal AI assistant for software engineering. Connect to Models Hub via BYOK using the customModels array in settings.json.

FluentRead - Open-Source Translation Plugin

An open-source browser translation plugin. Switch its built-in LLM engine to Models Hub to translate web pages.

Hermes Agent

An open-source terminal AI agent from Nous Research. Point its Custom Endpoint at Models Hub via the hermes model menu.

Immersive Translate

Browser translation extension. Point its custom AI service at Models Hub for page and subtitle translation.

LangBot - Instant Messaging Bot Development Platform

An open-source bot development framework for IM platforms. Add a model pointed at Models Hub for Feishu, WeChat, QQ, and more.

LunaTranslator - Open-Source GalGame Translator

An open-source GalGame translator. Add Models Hub's URL and key in its generic LLM interface to translate game text.

Memoh - Containerized AI Agent Platform

A containerized AI agent platform. Add the NewAPI provider type and enter Models Hub's URL and key.

nanobot

A lightweight, open-source personal AI agent. Edit providers and agents.defaults in config.json to connect to Models Hub.

NextChat

Open-source cross-platform chat client. Connect to Models Hub via a custom endpoint, hosted or self-deployed.

Oh My Pi

A terminal AI coding agent — add Models Hub as a custom provider by editing models.yml.

Open WebUI

Self-hosted chat UI. Add Models Hub under Admin → Connections to serve your whole team.

OpenClaw

Self-hosted AI assistant platform. Connect to Models Hub with one click or by editing providers.

OpenCode

An open-source terminal AI coding agent — connect with /connect or by editing opencode.json.

Pi

A minimal, highly extensible terminal coding framework — add a custom provider in models.json.

WorkBuddy

The CodeBuddy IDE assistant — add a custom model right from the settings UI.

Integration Guide

Most applications only need the following information to connect to Models Hub:

  1. API Base URL: The Models Hub service address, typically ending with /v1.
  2. API Key: The access token (API key) generated in the Models Hub console.
  3. Model name: The model ID you want to call.

For detailed configuration steps, refer to each application's documentation.

Compatibility tip

If a tool supports OpenAI-compatible, Custom Endpoint, or BYOK modes, you can usually fill in the Models Hub API Base URL, API Key, and model name in the corresponding fields to complete the connection.

On this page