> ## Documentation Index
> Fetch the complete documentation index at: https://docs.praison.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Messages CLI

> CLI commands for messages API and token counting

## Overview

Create messages and count tokens from the command line.

## Commands

### Create Message

```bash theme={"theme":{"light":"vitesse-light","dark":"vitesse-dark"}}
praisonai messages create "What is AI?"
```

### With System Prompt

```bash theme={"theme":{"light":"vitesse-light","dark":"vitesse-dark"}}
praisonai messages create "Write a poem" --system "You are a poet"
```

### With Model Selection

```bash theme={"theme":{"light":"vitesse-light","dark":"vitesse-dark"}}
praisonai messages create "Hello" --model gpt-4o
```

### Count Tokens

```bash theme={"theme":{"light":"vitesse-light","dark":"vitesse-dark"}}
praisonai messages count-tokens "Hello, how are you today?"
```

## Options

### messages create

| Option         | Short | Default                    | Description    |
| -------------- | ----- | -------------------------- | -------------- |
| `--model`      | `-m`  | claude-3-5-sonnet-20241022 | Model to use   |
| `--max-tokens` |       | 1024                       | Maximum tokens |
| `--system`     | `-s`  | None                       | System prompt  |

### messages count-tokens

| Option    | Short | Default     | Description            |
| --------- | ----- | ----------- | ---------------------- |
| `--model` | `-m`  | gpt-4o-mini | Model for tokenization |

## Examples

```bash theme={"theme":{"light":"vitesse-light","dark":"vitesse-dark"}}
# Create a message
praisonai messages create "Explain quantum computing briefly"

# Count tokens before sending
praisonai messages count-tokens "This is my prompt text"

# Use with system prompt
praisonai messages create "Write code" -s "You are a Python expert" -m gpt-4o
```