goviewmodels

goviewmodels

A terminal-based tool for discovering, browsing, and managing local AI model files across multiple AI applications.

Features

• Multi-App Support: Discovers models from Ollama, LM Studio, GPT4All, Jan, LocalAI, koboldcpp, text-generation-webui, and llama.cpp
• Multiple Formats: Supports GGUF (full metadata) and SafeTensors (basic) model formats
• Interactive TUI: Beautiful terminal user interface built with Bubble Tea for browsing models
• Non-Interactive CLI: Command-line tools for automation and scripting
• Centralized Repository: Hardlink-based model management to organize models in one location
• Rich Metadata: Extracts detailed model information including quantization, VRAM estimates, and parameters
• Cross-Platform: Works on Linux, macOS, and Windows

Installation

From Source

git clone https://github.com/JasonDoug/goviewmodels.git
cd goviewmodels
go build -o goviewmodels .

Using Go Install

go install github.com/JasonDoug/goviewmodels@latest

Quick Start

Launch the interactive TUI:

goviewmodels

List models in terminal:

goviewmodels list

Scan for models with optional linking:

goviewmodels scan [--link]

Usage

Interactive Mode

Run goviewmodels to launch the Terminal User Interface (TUI):

goviewmodels

Key bindings in TUI:

• j/k or ↑/↓: Navigate models
• Enter: View model details
• s: Scan for models
• l: Link selected model to repository
• /: Filter models by name/format/quant/source
• Tab: Cycle sort column
• v: Toggle VRAM estimation table (in detail view)
• ?: Show help
• q or Ctrl+C: Quit

Command Line Interface

List Models

# List all discovered models
goviewmodels list

# List models with additional directories
goviewmodels list --paths /path/to/custom/models

# Skip specific applications
goviewmodels list --skip ollama,lm-studio

Scan Models

# Scan for models
goviewmodels scan

# Scan and link all discovered models
goviewmodels scan --link

# Scan with additional directories
goviewmodels scan --paths /path/to/custom/models

# Skip specific applications
goviewmodels scan --skip ollama,lm-studio

Supported Applications

goviewmodels automatically discovers models from these AI applications:

Configuration

goviewmodels uses a JSON configuration file located at ~/.goviewmodels/config.json:

{
  "repo_path": "/home/user/.goviewmodels",
  "extra_paths": ["/path/to/additional/models"],
  "skip_apps": ["ollama"],
  "concurrency": 4,
  "prefer_hardlinks": true
}

Configuration options:

• repo_path: Directory for the centralized model repository
• extra_paths: Additional directories to scan for models
• skip_apps: List of applications to skip during scanning
• concurrency: Number of concurrent scanning workers
• prefer_hardlinks: Whether to prefer hardlinks over symlinks

Repository Management

goviewmodels creates a centralized model repository at ~/.goviewmodels/library by default. Models are hardlinked (or symlinked if cross-device) to this location for organized management.

The repository structure:

~/.goviewmodels/library/
├── ollama/
│   ├── llama3/
│   │   └── llama3.gguf
│   └── mistral/
│       └── mistral.gguf
├── lm-studio/
│   ├── TheBloke/
│   │   └── Llama-2-7B/
│   │       └── model.gguf
└── manifest.json

VRAM Estimation

goviewmodels provides VRAM estimates for GGUF models based on quantization type and context length. The detail view includes a comprehensive VRAM table showing memory requirements at different context lengths.

Building from Source

# Clone the repository
git clone https://github.com/JasonDoug/goviewmodels.git
cd goviewmodels

# Build
go build -o goviewmodels .

# Install
go install .

Contributing

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

1. Fork the repository
2. Create your feature branch (git checkout -b feature/AmazingFeature)
3. Commit your changes (git commit -m 'Add some AmazingFeature')
4. Push to the branch (git push origin feature/AmazingFeature)
5. Open a Pull Request

License

This project is licensed under the MIT License - see the LICENSE file for details.

Acknowledgments

• Bubble Tea - Terminal UI framework
• Bubble Table - Table component for Bubble Tea
• Cobra - CLI framework
• humanize - Formatting library