A lightweight desktop widget that provides instant access to local language models via Ollama engine.
Designed to be unobtrusive and efficient, PopChat acts as a quick overlay for your AI interactions without cluttering your workspace. It brings the power of local LLMs to your fingertips with a single keystroke.
- Instant Access: Toggle the chat window globally with a shortcut (default:
Ctrl+Shift+K) - Minimalist Design: A compact, floating UI that stays out of your way
- Local & Private: Fully powered by your local Ollama instance
- Configurable: Easily switch between models (e.g., Llama3, Mistral) via the settings menu)
- Ollama: Ensure Ollama is installed (you can check by
ollama list) - Model: Pull a model if you haven't already (e.g.,
ollama pull llama3)
PopChat is build with C++ and Qt6.
Remember to provide the path to Qt library in .env file (look .env.example).
- Launch the application
- Click the settings icon to select or change the active model
- Type your prompt and hit enter
- Press Ctrl+Shift+k to show or hide the widget
Distributed under the MIT License. See LICENSE for more information.