Llaminal is the ultimate CLI companion for your local Ollama instance. Stop wrestling with raw API calls and curl commands. Embrace the terminal with style.

Featuring a cyberpunk aesthetic, Llaminal gives you a powerful, persistent, and multimodal interface to your local LLMs.

• 🗣️ Interactive Chat: Full REPL with history, slash commands, and session management.
• 👁️ Multimodal Support: Drag-and-drop images into your terminal chat to use Vision models.
• 📚 RAG & Context: Pipe files directly into ask or use /add to read entire directories into context.
• 💾 Sessions: Save, list, and reload your conversations anytime.
• 🎨 Rich UI: Beautiful markdown rendering, tables, and spinners.
• 🛠️ Model Ops: Pull, show, and delete models without leaving the tool.
• 🔄 Resilient: Automatic retry on transient network errors with helpful diagnostics.

# Clone the repository
git clone <repo-url>
cd Llaminal

# Install locally
python3 -m venv venv
source venv/bin/activate
pip install -e .

1. The "One-Shot" Ask Perfect for quick scripts or piped debugging.

# Simple question
llaminal ask "What is the capital of Peru?"

# Debugging a log file
cat error.log | llaminal ask "Explain this error and fix it"

2. The Interactive Chat Your main command center.

llaminal chat

• Type /help to see all commands.
• Type /image ./path/to/img.png to attach an image.
• Type /add ./src to read your codebase.

3. Model Management

llaminal list
llaminal pull tinyllama
llaminal show llama3.2

Control Llaminal via environment variables:

If you love Llaminal, consider supporting the development!

Built with ❤️ by a developer who loves the terminal.