Get Started with Your AI Power Box

Start using your Personal AI Power Box in minutes.

Download

CLI Installation (macOS and Linux)

curl -fsSL https://raw.githubusercontent.com/nile-agi/delta/main/install.sh | bash

Quick Start

Get started in 3 steps:

Install Delta

Choose your platform and run the installation command:

# macOS\nbrew tap nile-agi/delta-cli && brew install --HEAD nile-agi/delta-cli/delta-cli\n\n# Linux (Debian/Ubuntu)\ncurl -fsSL https://raw.githubusercontent.com/nile-agi/delta/main/packaging/linux/install-deb.sh | sudo bash\n\n# Windows\nwinget install DeltaCLI.DeltaCLI

Run Delta

Start your AI Power Box - it will auto-download the default model:

delta

This auto-downloads qwen2.5:0.5b (~400MB) if not installed, then starts interactive mode.

Start Using

Ask questions, analyze documents, or start the web UI:

# Ask a question\ndelta "What is artificial intelligence?"\n\n# Start web server\ndelta server\n# Then open http://localhost:8080

Command Reference

Commands to control your AI Power Box

Basic Commands

Interactive mode:
```
delta
```
Auto-downloads default model if not installed
Single query:
```
delta "What is the capital of France?"
```
List local models:
```
delta --list-models
```
Download a model:
```
delta pull qwen2.5:0.5b
```
Start web server:
```
delta server
```

Advanced Features

Use specific model:

delta --model llama3.1:8b "Your question here"

With GPU acceleration:

delta --model qwen2.5:0.5b --gpu-layers -1 "Your question"

Multimodal mode:

delta --multimodal "Describe this image"

Custom context size:
```
delta --ctx-size 4096 "Your question"
```
Check for updates:
```
delta --check-updates\ndelta --update
```

Complete Command List

Model Management

delta pull <model>          # Download a model

delta remove <model>        # Remove a model

delta --list-models         # List local models

delta --list-models --available  # List available models

Usage

delta                          # Interactive mode

delta "Your question"          # Single query

delta --model <model> <query>  # Use specific model

delta server                   # Start web server

delta server --port 8081       # Custom port

Options

delta -h, --help              # Show help

delta -v, --version          # Show version

delta -t, --tokens <N>      # Max tokens (default: 512)

delta -T, --temperature <F> # Temperature (default: 0.8)

delta -c, --ctx-size <N>     # Context size (default: 2048)

delta -g, --gpu-layers <N>  # GPU layers (default: 0, -1 for all)

delta --multimodal           # Enable multimodal mode

Interactive Mode Commands

In interactive mode, use slash commands to control your AI Power Box:

/download <model>

Download a model

/remove <model>

Remove a model

/list

List local models

/available

List available models

/use <model>

Switch to another model

/tokens <N>

Set max tokens (default: 512)

/temperature <F>

Set temperature (default: 0.8)

/gpu-layers <N>

Set GPU layers (default: 0, -1 for all)

/multimodal

Toggle multimodal mode

/server

Start web dashboard

/help

Show all commands

exit, quit

Exit interactive mode

Ready to Power Up Your AI Box?

Start using your Personal AI Power Box today

View Installation Guide View on GitHub