Switch Configuration

Opening the configuration interface

Menu bar icon

macOS / Linux: click the menu bar icon → "Configuration".

System tray

Windows: right-click the tray icon → "Open Configuration".

Keyboard shortcut

Ctrl+Shift+S (Win/Linux) / Cmd+Shift+S (macOS).

---

Adding a model provider

In the "ProviderProviderA model provider, i.e. an upstream model service (a cloud API or a local inference service). Switch supports configuring multiple Providers and routing by rule." tab → "Add Provider", enter the provider name + API Base URL + API Key:

Provider	API Base URL	API Key
Lurus API Recommended	https://hub.lurus.cn/v1	Lurus Key (starts with sk-); for models, click "Auto-detect"
OpenAI	https://api.openai.com/v1	sk-... (official)
Anthropic	https://api.anthropic.com/v1	sk-ant-...
Ollama (local)	http://localhost:11434/v1	(leave empty)

---

Routing

Configuring routing rules

Define which request goes to which provider; unmatched requests go to the default provider (default → Lurus API).

Route by model name: gpt-* → OpenAI; claude-* → Anthropic; deepseek-* / * (others) → Lurus API; llama* → Ollama. JSON:

{ "rules": [
    { "pattern": "gpt-*", "provider": "OpenAI" },
    { "pattern": "claude-*", "provider": "Anthropic" },
    { "pattern": "llama*", "provider": "Ollama" },
    { "pattern": "*", "provider": "Lurus API" }
  ] }

Route by application (advanced): set different routes for different local applications:

{
  "app_rules": [
    {
      "app": "cursor",
      "default_provider": "Lurus API",
      "model_override": "deepseek-reasoner"
    },
    {
      "app": "continue",
      "default_provider": "Ollama"
    }
  ]
}

---

Proxy port configuration

"General" → "Listen Port", default 11434. If there is a port conflict (e.g. with Ollama), change it to another port (such as 11435), and update the application side accordingly with base_url=http://localhost:11435/v1 (api_key can be anything; Switch uses the configured provider key).

Port conflict

The default port 11434 is the same as Ollama’s default port. When both run on the same machine, change Switch to another port (such as 11435), and update the application-side base_url accordingly.

---

Full configuration file

Switch configuration is saved at:

Platform	Path
Windows	%APPDATA%\LurusSwitch\config.json
macOS	~/Library/Application Support/LurusSwitch/config.json
Linux	~/.config/LurusSwitch/config.json

Full config.json example:

{
  "listen_port": 11434,
  "log_requests": true,
  "start_on_login": true,
  "providers": [
    {
      "name": "Lurus API",
      "base_url": "https://hub.lurus.cn/v1",
      "api_key": "sk-your-lurus-key",
      "enabled": true
    },
    {
      "name": "Ollama",
      "base_url": "http://localhost:11434/v1",
      "api_key": "",
      "enabled": true
    }
  ],
  "routing": {
    "default_provider": "Lurus API",
    "rules": [
      { "pattern": "llama*",  "provider": "Ollama" },
      { "pattern": "qwen*",   "provider": "Ollama" },
      { "pattern": "*",       "provider": "Lurus API" }
    ]
  }
}

---

Verifying the configuration

The "Status" tab displays: the connection status of each provider (green = OK, red = failed), the currently active routing rules, and recent request logs. Command-line verification (a normal JSON response means success):

curl http://localhost:11434/v1/chat/completions \
  -H "Content-Type: application/json" \
  -d '{"model":"deepseek-chat","messages":[{"role":"user","content":"ping"}]}'

---

Next steps

Next steps

Cost monitoringMCP serversTeam configuration