golms

A Local Model Server Interface written in Go for managing and interacting with LLM model servers. This was initially built as a challenge for myself to code a useful tool in Go during a 6-hour flight with limited internet access (Chang Min).

Overview

golms is a CLI tool that provides a unified interface for managing and chatting with local LLM models across different model servers. It currently supports:

• MLX LM - Apple Silicon optimized model server
• Ollama - Cross-platform model server (WIP)

Features

• 📋 List available LLMs and model servers
• 🔌 Connect to model servers and chat with LLMs
• 🎯 Automatic model server discovery and management
• 💬 Interactive chat interface with styled TUI
• 🎨 Beautiful terminal UI with color-coded messages and status indicators
• 🚀 Support for multiple model server backends

Prerequisites

Make sure you have the following installed:

• Go (1.25.1 or later)
• Model Server (at least one of the supported servers):
  • MLX LM for Apple Silicon
  • Ollama for cross-platform support
• LLM Models downloaded in ~/golms/<model_server>/ directories
  • For example: ~/golms/mlx_lm/ or ~/golms/ollama/

Installation

Using go install (Recommended)

go install github.com/changminbark/golms@latest

This will install the latest version to $GOPATH/bin. Make sure $GOPATH/bin is in your $PATH.

From Source

# Clone the repository
git clone https://github.com/changminbark/golms.git
cd golms

# Build the binary
make build

# Or install to $GOPATH/bin
make install

Build Modes

Default Mode (uses cache, ignores vendor/)

go build

Uses dependencies from ~/go/pkg/mod/

Vendor Mode (uses vendor/)

go build -mod=vendor

Uses dependencies from ./vendor/

Usage

List Available LLMs and Model Servers

golms list

This will show all available LLMs organized by model server and list all installed model servers.

List Supported Model Servers

golms servers

Shows all model servers that golms supports.

Connect to a Model Server and Chat

golms connect

This will:

1. Prompt you to select a model server
2. Show available LLMs for that server
3. Start the model server (if not already running)
4. Connect you to an interactive chat session

Project Structure

golms/
├── cmd/
│   └── root.go              # CLI commands and handlers
├── pkg/
│   ├── client/              # Client implementations for model servers
│   │   ├── client.go
│   │   ├── mlx_lm.go
│   │   └── ollama.go
│   ├── constants/           # Constants and configurations
│   │   └── model_server.go
│   ├── discovery/           # Model and server discovery
│   │   └── discovery.go
│   ├── server/              # Model server management
│   │   ├── manager.go
│   │   ├── mlx_lm.go
│   │   └── ollama.go
│   ├── ui/                  # Terminal UI styles and formatting
│   │   └── styles.go
│   └── utils/               # Utility functions
│       ├── clean.go
│       └── clean_test.go
├── main.go                  # Application entry point
├── go.mod
├── go.sum
├── Makefile
└── README.md

Development

Building

# Build the binary
make build

# Build with vendor dependencies
make build-vendor

Testing

# Run all tests
make test

# Run tests with coverage
make test-coverage

Code Quality

# Format code
make fmt

# Run go vet
make vet

# Run all checks (fmt, vet, test)
make check

Cleaning

# Remove binary and clean build cache
make clean

Commands Reference

Command	Description
golms	Show usage information
golms list	List all available LLMs and model servers
golms servers	List all supported model servers
golms connect	Connect to a model server and start chatting with an LLM

Configuration

Models should be placed in the following directory structure:

~/golms/
├── mlx_lm/
│   ├── model-1/
│   └── model-2/
└── ollama/
    ├── model-3/
    └── model-4/

Each model directory should contain the necessary model weights and configuration files required by the respective model server.

License

See LICENSE file for details.

Contributing

Contributions are welcome! Please feel free to submit a Pull Request.