gerbil/README.md
2025-08-22 22:37:21 -07:00

4.4 KiB

Friendly Kobold

A modern desktop app for running Large Language Models locally. FriendlyKobold Icon

Core Features

  • Run LLMs locally with a modern, intuitive interface powered by KoboldCpp
  • Full Linux Wayland support for seamless integration with modern Linux desktops
  • Automatic updates - Download and keep your KoboldCpp binary up-to-date effortlessly
  • Smart process management - Prevents runaway background processes and system resource waste
  • Optimized performance - Automatically unpacks binaries for faster operation and reduced memory usage (up to ~4GB less RAM)
  • Image generation support - Built-in presets for Flux and Chroma image generation workflows
  • Adaptive theming - Light, dark, and system theme modes that automatically follow your OS preferences
  • Privacy-focused - Everything runs locally on your machine, no data sent to external servers

Installation

Download the latest release for your platform from the GitHub Releases page:

  • Windows: Friendly-Kobold-X.X.X.exe (portable executable)
  • macOS: Friendly-Kobold-X.X.X.dmg (disk image)
  • Linux: Friendly-Kobold-X.X.X.AppImage (portable application)

Linux - AppImage

  1. Download the .AppImage file
  2. Make it executable: chmod +x Friendly-Kobold-X.X.X.AppImage
  3. Run it: ./Friendly-Kobold-X.X.X.AppImage

Linux - AUR (Arch Linux)

For Arch Linux users, install from the AUR using your preferred AUR helper:

# Using yay
yay -S friendly-kobold

# Using paru
paru -S friendly-kobold

# Manual installation
git clone https://aur.archlinux.org/friendly-kobold.git
cd friendly-kobold
makepkg -si

The AUR package automatically handles installation, desktop integration, and system updates. This is the ideal way to run Friendly Kobold on Linux.

Screenshots

Download & Setup

Download Interface

Model Launch Configuration

Launch Configuration

Terminal Output

Terminal Interface

Text Generation

Text Story Generation

Image Generation

Image Generation

Windows ROCm Support

There is ROCm Windows support maintained by YellowRoseCx in a separate fork. Unfortunately it does not properly support unpacking, which would greatly diminish its performance and provide a poor UX when used alongside this app. For Friendly Kobold to work with this fork, this issue must be fixed first.

Note that this build is not important as the modern day Vulkan backend matches or even surpasses ROCm in terms of LLM performance for most cases.

Future features

Not all koboldcpp features have currently been ported over the UI. As a workaround one may use the "Additional arguments" on the "Advanced" tab of the launcher to provide additional command line arguments if you know them.

For Local Dev

Prerequisites

  • Volta - JavaScript tool manager (installs correct Node.js + Yarn versions automatically)

Setup

  1. Install Volta (if not already installed):

    curl https://get.volta.sh | bash
    
  2. Clone the repository

  3. Install dependencies (Volta will auto-install the correct Node.js and Yarn versions):

    yarn
    
  4. Start the development server:

    yarn dev
    

Future considerations

It would make a lot of sense to transition this project to Tauri from Electron. The app size should drop from ~110MB to ~10MB; however, users on obsolete OSes (with outdated WebViews) will very likely encounter issues. In addition, I would need to learn Rust to rewrite the BE (Electron main code), but at least we can re-use all the React code. The app would be much smaller, faster and memory efficient, but not work for some users. I think it's a worthy tradeoff.

License

AGPL v3 License - see LICENSE file for details