Chanh 🍋

A VS Code extension that downloads, hosts, and runs the Lemonade Server embeddable binary (lemond) locally, enabling AI chat with local LLMs directly in VS Code.

⚠️ Early stage: The 0.0.1 release focuses primarily on setup: downloading and hosting the binary, plus basic server and model management. More functionality is planned for upcoming releases. Testing is still limited, and development has so far been done only on Windows. Other platforms and edge cases may not work as expected. Feedback and issue reports are welcome.

Features

• Managing/downloading lemonade models
• Chatting integration

Chat Commands

The @chanh chat participant supports the following slash commands:

• /fix - Generate a fix for the selected code
• /explain - Explain the selected code

Configuration

Setting	Default	Description
chanh.chatModel	""	Model to use for chat (leave empty to be prompted)
chanh.customServerUrl	(unset)	Custom Lemonade Server URL used when chanh.targetServer is "custom" (e.g., http://localhost:13305)
chanh.embeddedPort	8000	Port for the embedded lemond server
chanh.maxLoadedModels	(unset)	Maximum number of loaded models. Use -1 for unlimited.
chanh.standalonePort	13305	Port for the standalone Lemonade Server
chanh.targetServer	"standalone"	Which Lemonade Server to use: standalone, embedded, or custom

Server Mode Selection

You can choose which Lemonade server the extension uses by setting chanh.targetServer (Chanh: Select Server lets you pick from the command palette, which updates this setting automatically):

Value	Behavior
standalone (default)	Connects to an existing system-installed standalone Lemonade Server.
embedded	Always starts the bundled lemond binary.
custom	Connects to the URL in chanh.customServerUrl.

Release Notes

0.0.2

• Add icon

0.0.1

• Lemonade management treeview
• Chat integration

Enjoy!