https://github.com/redromnon/picochat
A minimal and fast LLM chat app with temporary session
https://github.com/redromnon/picochat
chatbot llm qt
Last synced: 4 days ago
JSON representation
A minimal and fast LLM chat app with temporary session
- Host: GitHub
- URL: https://github.com/redromnon/picochat
- Owner: redromnon
- Created: 2026-02-21T09:45:05.000Z (5 months ago)
- Default Branch: master
- Last Pushed: 2026-02-21T15:40:38.000Z (5 months ago)
- Last Synced: 2026-02-21T17:55:12.859Z (5 months ago)
- Topics: chatbot, llm, qt
- Language: Python
- Homepage:
- Size: 10.7 KB
- Stars: 0
- Watchers: 0
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# PicoChat
A minimalist LLM chat application that only keeps one *temporary* chat session running. A new session can be created using a simple shortcut `Ctrl+Shift+D`. Built using Python and PyQt (PySide6).
Ideal for short and quick tasks. Inspired by Firefox Focus.
### 🤔 Why build this?
I wanted a quick and efficient way to execute simple tasks like proofreading, minor bug/code explanations, etc., using language models. Things I don't mind storing for later use.
Utilising small language models running locally on my laptop couldn't have been a better solution.
Another reason was that I wanted to vibecode a simple app whilst testing Google's Antigravity.
## Installation & Running
- If you're a **Linux** user, download the pre-built binary from the [Releases](https://github.com/redromnon/picochat/releases/tag/v1.0) section. All binaries are built on a Fedora machine.
- If you're using Windows or macOS, you'll need to build the app yourself.
You will need a language model already running on your system using an inference engine like LMStudio, Ollama, etc. Then insert the OpenAI-compatible model endpoint in the app's top bar.