Skip to main content

A GTK graphical interface for chatting with large language models (LLMs)

Project description

GTK LLM Chat

A GTK graphical interface for chatting with Large Language Models (LLMs).

screenshot

Key Features

  • Simple and easy-to-use graphical interface built with GTK

  • Support for multiple conversations in independent windows

  • Integration with python-llm for chatting with various LLM models

  • Modern interface using libadwaita

  • Support for real-time streaming responses

  • Message history with automatic scrolling

  • Windows installer, Linux AppImage, and Macos bundles available!

  • Markdown rendering of the responses

  • Sidebar Navigation: Modern sidebar for model/provider selection, parameters, and settings.

  • Model Parameters: Adjust temperature and system prompt per conversation.

  • API Key Management: Banner with symbolic icons for setting/changing API keys per provider.

  • Keyboard Shortcuts:

    • F10: Toggle sidebar
    • F2: Rename conversation
    • Escape: Minimize window
    • Enter: Send message
    • Shift+Enter: New line in input
    • Ctrl+W: Delete the current conversation
    • Ctrl+M: Open model selector
    • Ctrl+S: Edit system prompt
    • Ctrl+N: New conversation window
  • Conversation Management: Rename and delete conversations.

  • Tray Applet: Use a system tray applet for quick access to recent conversations.

  • Error Handling: Clear error messages displayed in the chat.

  • Dynamic Input: The input area dynamically adjusts its height.

Gtk-LLM-Chat is a graphical frontend for the command-line llm utility. Just as llm integrates large language models into the command line interface, Gtk-LLM-Chat aims to bring that same power to the desktop environment. Its goal is to provide intuitive affordances and seamless integration for using LLMs in everyday tasks — all while remaining convenient, lightweight, and transparent in its behavior.

Installation

Downloadable application bundles

While the command line is fun in every operating system, Gtk-LLM-Chat also offers prepackaged binary application bundles for all three major operating sytems: Windows installers, Linux Appimages and Macos Application Bundles are available in our releases section.

An effort has been made to support desktop integration across systems but your mileage may vary - as the Gtk tools are still maturing outside of the GNU/Linux ecosystem.

As an llm plugin

Playing with LLMs in the command line is fun! I recommend you to install llm and play around with it to size up the possibilities. Gtk-LLM-Chat can be installed as a plugin extension for llm itself, thus extending the possibilities of llm with some graphical features. Not all features of llm are exposed yet.

pipx install llm               # required by gtk-llm-chat
llm install gtk-llm-chat

You may want to copy the provided .desktop files to your ~/.local/share/applications/ folder. A welcome assistant will do this in the future for you.

System Requirements

  • llm (when installing as an llm plugin)
  • Python 3.8 or higher
  • GTK 4.0
  • libadwaita
  • libayatana-appindicator (on linux)

These dependency installation instructions are collected here for reference only:

 # fedora: # sudo dnf install cairo-devel object-introspection-devel gtk4-devel pkgconf-pkg-config gcc redhat-rpm-config
 # debian: # sudo apt install libgtk-4-1 python3-gi python3-gi-cairo libadwaita-1-0 libayatana-appindicator3
 # arch: # sudo pacman -S python-gobject gtk4
 # windows (msys2): # pacman -S mingw-w64-$(uname -m)-gtk4 mingw-w64-$(uname -m)-python-pip mingw-w64-$(uname -m)-python3-gobject mingw-w64-$(uname -m)-libadwaita mingw-w64-x86_64-python3-pillow
 # macos (homebrew): # brew install pygobject3 gtk4 adwaita-icon-theme libadwaita

Usage

Running the Application

To start the applet (system tray mode):

llm gtk-applet

To start a single chat window:

llm gtk-chat

Optional arguments:

llm gtk-chat --cid CONVERSATION_ID   # Continue a specific conversation
llm gtk-chat -s "System prompt"      # Set system prompt
llm gtk-chat -m model_name           # Select specific model
llm gtk-chat -c                      # Continue last conversation

Development

To set up the development environment:

git clone https://github.com/icarito/gtk-llm-chat.git
cd gtk-llm-chat
python -m venv venv
source venv/bin/activate
pip install -e .

Shoulders of giants

This project is made possible thanks to these great components, among others:

License

GPLv3 License - See LICENSE file for more details.

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

gtk_llm_chat-3.2.0.tar.gz (1.1 MB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

gtk_llm_chat-3.2.0-py3-none-any.whl (164.3 kB view details)

Uploaded Python 3

File details

Details for the file gtk_llm_chat-3.2.0.tar.gz.

File metadata

  • Download URL: gtk_llm_chat-3.2.0.tar.gz
  • Upload date:
  • Size: 1.1 MB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.1.0 CPython/3.12.9

File hashes

Hashes for gtk_llm_chat-3.2.0.tar.gz
Algorithm Hash digest
SHA256 6f5141f631a9f67095dda93befbe82cf0787dfc1b7115ac6e523183df15aadae
MD5 9ce51f6369879bcec2437230a0636a9b
BLAKE2b-256 4abcac93731ca9d43dd0134f45957db35087c0787e83f9c94b4281787f9f210e

See more details on using hashes here.

File details

Details for the file gtk_llm_chat-3.2.0-py3-none-any.whl.

File metadata

  • Download URL: gtk_llm_chat-3.2.0-py3-none-any.whl
  • Upload date:
  • Size: 164.3 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.1.0 CPython/3.12.9

File hashes

Hashes for gtk_llm_chat-3.2.0-py3-none-any.whl
Algorithm Hash digest
SHA256 836dfd4f5ef8da8959798fc7b609b2012c30f444b8ff7155c3f89d683db4149b
MD5 36cc4fbcaf3a1958df739a371cf1d9c9
BLAKE2b-256 1f078c59d648e4f34527217c2ba8bf91712971c8562a9f57ea32c99324a312e0

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page