Multi-modal AI assistant with desktop GUI, web interface, and CLI. Runs locally on Ollama with support for small and large language models.
Multiple interfaces for different use cases
| Command | Version | Mode |
|---|---|---|
python ui2.py | V2 | Desktop GUI (Tkinter, animated HUD) |
python web_v2.py | V2 | Web server on :9090 |
python jarvis_v2.py | V2 | CLI with native tool-calling agent |
python ui2_small.py | V2 Small | Desktop GUI for 1B-3B models (green theme) |
python web_v2_small.py | V2 Small | Web server on :9091 for small models |
python jarvis_v2_small.py | V2 Small | CLI agent for small models |
Interact via voice commands or text input. Multi-modal conversation with context retention.
Animated Tkinter HUD interface with real-time status indicators and chat history.
Browser-based chat interface accessible from any device on your network.
V2 uses Ollama's native tool_calls for multi-step reasoning with web search and file operations.
Optimized for Pi 4 with small model support. Qwen2.5 1.5B is the recommended sweet spot.
Read, write, and manage files directly through the assistant. Multi-step task execution.