Skip to content

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

3 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

🦦 Navi — AI Desktop Pet

A pixel-art desktop companion powered by a local AI (Ollama). Navi lives on your desktop, reacts to what you're doing, and chats with you in real time.

⚠️ Windows only. macOS and Linux are not supported at this time.


Requirements

  • Windows 10 / 11
  • Python 3.11+
  • Ollama installed and running
  • At least 8GB RAM (16GB recommended for larger models)

Installation

1. Clone the repository

git clone https://github.com/RedSky09/Navi.git
cd Navi

2. Install Python dependencies

pip install -r requirements.txt

3. Install Ollama

Download and install from https://ollama.com, then run:

ollama serve

4. Download the AI model

ollama pull llama3.2

This will download ~2GB. Only needs to be done once.

5. Run the app

python main.py

Usage

  • Left-click + drag — move Navi around your screen
  • Right-click — open the menu (Talk, Play, Sleep, Settings, Quit)
  • Talk — open the chat bubble and type a message
  • Play — play a game of Checkers against Navi
  • Sleep — put Navi to sleep (wake up via menu)
  • Settings — change name, personality, language, and more

AI & Chat

Navi uses a local AI model via Ollama to generate responses. Replies are intentionally kept short (1–2 sentences) to match the pet's personality.

Limitations:

  • Responses may occasionally be cut off — this is by design to keep Navi feeling lightweight and fun, not like a full chatbot.
  • Memory is limited to recent conversations within a session, plus a summary of past sessions.
  • Response quality depends on the model you use. llama3.2 is recommended for the best balance of speed and quality.

Troubleshooting:

  • "Ollama is not running!" — make sure ollama serve is running in the background.
  • "Model not ready!" — run ollama pull llama3.2 in your terminal.
  • Navi is silent on startup — Ollama may still be loading. Wait a few seconds and right-click to open the menu.

Configuration

Settings are saved to config.json. You can edit this file directly or use the in-app Settings panel.

Key Default Description
pet_name Navi Your pet's name
pet_type otter Pet species
personality playful playful, calm, sassy, shy
model llama3.2 Ollama model to use
memory_limit 10 Max conversation turns kept in memory
user_name `` Your name (Navi will call you by name)
language AUTO AUTO, ENGLISH, INDONESIA
pet_size 200 Pet size in pixels

Disclaimer

Navi runs entirely on your local machine via Ollama. Response quality may vary depending on the model and your hardware. The developer is not responsible for any content generated by the AI model.


License

MIT License — feel free to use, modify, and share.


Credits

Asset packs used in this project:

Asset Author Link
Otter Sprite Pack RiLi_XL itch.io
Pixel Checkers - Free asset pack dani maccari itch.io
Sprout Lands - UI Pack Cup Nooble itch.io

About

A pixel-art desktop companion powered by a local AI (Ollama). Navi lives on your desktop, reacts to what you're doing, and chats with you in real time

Topics

Resources

Stars

2 stars

Watchers

0 watching

Forks

Releases

Packages

Contributors

Languages