All tools

yetone

magpie + LLMoxy

Pick models for all your agents in one place. Claude Code, Codex, OpenCode, and others each use their own model, routed through one local gateway.

OpenAI-compatiblehttps://en.llmoxy.com/v1
Anthropic-compatiblehttps://en.llmoxy.com
Gemini-compatiblehttps://en.llmoxy.com/v1beta

Quick Start

  1. 1

    Install magpie

    Install with the official script. It sets up both the desktop app and the magpie command. On Windows, download it from the official website.

    Go to downloads
    curl -fsSL https://usemagpie.ai/install.sh | sh
    BASH
  2. 2

    Create an API key

    Create a token in the Console, pick the group you want to use, and copy the key.

    Create a token
  3. 3

    Add LLMoxy as a provider

    The button below opens magpie with the name and API URLs already filled in. Paste your key and click "Add". To add it manually instead, go to "Providers" → "+ Add provider" → "Custom · any compatible URL" and enter the URLs below.

    Add to magpie
    https://en.llmoxy.com/v1
    URL
  4. 4

    Pick a model for each agent

    On the "Agents" tab, click an agent's model and pick one that starts with llmoxy/. If a model isn't listed, type its ID. You can also use the command line:

    magpie claude llmoxy/claude-opus-5-5
    magpie codex llmoxy/gpt-5.6-sol
    magpie ls
    BASH
  5. 5

    Get started

    Agents read their config only at launch, so start a new session to use the change. For Codex, also restart the app and any open codex sessions.

Tips

  • One token works for every agent magpie supports, including Claude Code, Codex, and OpenCode. No need to edit each config file.
  • You don't need to export OPENAI_BASE_URL or ANTHROPIC_BASE_URL yourself. magpie writes the gateway URL into each agent's own config.
  • Keep magpie running. It stays in the menu bar after you close the window. magpie serve runs only the gateway, and magpie daemon install installs it as a service that starts at login.
  • To restore an agent's original config, run magpie <agent> default.

FAQ

Nothing happens when you click the button?
One-click import requires magpie 0.1.8 or later. Install it and open it once first. You can also add the provider manually as described above.
Getting a 502 error?
A proxy app such as Surge or Clash is intercepting 127.0.0.1:3425. Add a DIRECT rule for 127.0.0.0/8 and place it before your other rules.
Agent still using the old model after you switched?
Running sessions keep the model they started with. Start a new session.

Other tools

Claude

Anthropic

Claude

Anthropic's AI assistant and coding agent: Claude Desktop, plus Claude Code in the terminal and in VS Code.

DesktopTerminalVS Code
Codex

OpenAI

ChatGPT (Codex)

OpenAI's coding agent. The ChatGPT desktop app, the CLI, and the IDE extension share one config.toml, so one setup covers all three.

DesktopTerminalIDE extension
Antigravity

Google

Antigravity

Google's coding agent, the successor to Gemini CLI. For now, only the CLI lets you change the API URL.

View guide
opencode

Anomaly

OpenCode

Open-source coding agent with a desktop app and a terminal UI. You can configure multiple model providers.

DesktopTerminal
DeepSeek

DeepSeek

DeepSeek Harness

DeepSeek's open-source agent framework, with a desktop app, a web UI, and the dsh command line. It connects to OpenAI- or Anthropic-compatible model services.

DesktopWebTerminal
Cursor

Anysphere

Cursor

AI editor built on VS Code. Override the OpenAI URL to route OpenAI models in chat through LLMoxy.

View guide
GithubCopilot

GitHub

GitHub Copilot

The Copilot app, Copilot CLI, and VS Code all support bring your own key (BYOK), with no Copilot subscription required.

DesktopTerminalVS Code
TRAE

ByteDance

Trae

ByteDance's AI IDE. Both the China and international versions support custom models.

View guide
Cline

Cline

Cline

Open-source coding agent with a full agent loop, available as a command-line tool and as VS Code and JetBrains extensions.

TerminalIDE extension
Qoder

Alibaba

Qoder

Alibaba's AI coding product. The Qoder desktop app and Qoder CLI support custom models; Qoder IDE doesn't yet.

DesktopTerminal
CodeBuddy

Tencent

CodeBuddy

Tencent's AI coding tools. Both the command-line CodeBuddy Code and CodeBuddy IDE support OpenAI-compatible custom models.

TerminalIDE
WorkBuddy

Tencent

WorkBuddy

Tencent's desktop AI workspace for office and coding tasks. It supports custom model services.

View guide
CherryStudio

CherryHQ

Cherry Studio

Cross-platform desktop AI client. Add a provider to fetch the list of available models in one click.

View guide

Chatbox AI

Chatbox

AI chat client for desktop and mobile. Add a model provider in settings to connect.

View guide
Dify

LangGenius

Dify

Open-source LLM app development platform. Connect it as OpenAI-API-compatible to build apps and workflows.

View guide
n8n

n8n

n8n

Workflow automation platform. Enter the LLMoxy URL in an OpenAI credential to call models.

View guide