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

# Image Generation

> Create images with AI using natural language descriptions

## Overview

EnConvo's Image Generation feature lets you create images directly from text descriptions. Generate artwork, illustrations, mockups, and visual content without leaving your workflow.

## Supported Models

| Model                | Provider                     | Best For                                                           |
| -------------------- | ---------------------------- | ------------------------------------------------------------------ |
| **GPT Image 2**      | OpenAI / OpenAI Subscription | High quality image generation through API or OAuth account support |
| **DALL-E 3**         | OpenAI                       | High quality, accurate prompts                                     |
| **DALL-E 2**         | OpenAI                       | Fast generation, variations                                        |
| **xAI Imagine**      | xAI                          | Fast prompt-driven image generation through xAI                    |
| **Stable Diffusion** | Local/API                    | Privacy, customization                                             |
| **Midjourney**       | API                          | Artistic, stylized images                                          |

## Generating Images

### From SmartBar

1. Open SmartBar with `⇧⌘D`
2. Type `/image` followed by your description
3. Press Enter to generate

### Example Prompts

```
/image A serene mountain lake at sunset with snow-capped peaks
/image Minimalist logo for a tech startup, blue and white colors
/image Cute cartoon cat sitting on a stack of books
/image Professional headshot of a friendly robot in a business suit
```

### Advanced Prompts

Include style and technical specifications:

```
/image [subject], [style], [details], [lighting], [mood]

Example:
/image A cozy coffee shop interior, watercolor style, 
warm lighting, vintage furniture, peaceful atmosphere
```

## Image Options

### Size Options

| Size                      | Use Case                      |
| ------------------------- | ----------------------------- |
| **Square (1024×1024)**    | Social media, general purpose |
| **Landscape (1792×1024)** | Banners, presentations        |
| **Portrait (1024×1792)**  | Mobile, Pinterest             |

### Quality Settings

| Quality      | Description                   |
| ------------ | ----------------------------- |
| **Standard** | Fast generation, good quality |
| **HD**       | Higher detail, slower         |
| **Draft**    | Quick preview, lower quality  |

### Style Presets

| Preset       | Description                  |
| ------------ | ---------------------------- |
| **Natural**  | Photorealistic style         |
| **Vivid**    | Enhanced colors and contrast |
| **Artistic** | Creative, stylized           |
| **Anime**    | Japanese animation style     |

## Working with Generated Images

### After Generation

* **Preview**: View full-size image
* **Save**: Download to your Mac
* **Copy**: Copy to clipboard
* **Edit**: Modify with new prompt
* **Variations**: Generate similar images

### Edit & Refine

Use follow-up prompts to refine:

* "Make the background darker"
* "Add more contrast"
* "Change the style to watercolor"
* "Remove the person on the left"

### Variations

Generate variations of an image you like:

1. Click "Variations" on any generated image
2. EnConvo creates similar images with subtle differences
3. Choose your favorite

## Choosing A Provider

| Provider                     | Choose it when                                                                |
| ---------------------------- | ----------------------------------------------------------------------------- |
| **OpenAI API**               | You want direct API billing and access to OpenAI image models                 |
| **OpenAI Subscription**      | You want to use eligible image features through your connected OpenAI account |
| **xAI**                      | You want to generate images with xAI Imagine                                  |
| **Local / Stable Diffusion** | You want local or self-hosted image generation                                |

<Note>
  Model availability depends on your provider account and EnConvo version. If a
  new model does not appear, update EnConvo, validate the provider credential, and
  reopen the command settings.
</Note>

## Prompt Engineering

### Effective Prompt Structure

```
[Subject] + [Style] + [Environment] + [Lighting] + [Technical]
```

### Style Keywords

| Category        | Keywords                                      |
| --------------- | --------------------------------------------- |
| **Art Styles**  | oil painting, watercolor, digital art, sketch |
| **Photography** | portrait, landscape, macro, aerial            |
| **Aesthetic**   | minimalist, vintage, futuristic, retro        |
| **Mood**        | dramatic, peaceful, energetic, mysterious     |

### Technical Modifiers

| Modifier          | Effect                 |
| ----------------- | ---------------------- |
| `4K`, `8K`        | Higher detail          |
| `cinematic`       | Movie-like composition |
| `studio lighting` | Professional lighting  |
| `bokeh`           | Blurred background     |
| `rule of thirds`  | Better composition     |

## Use Cases

<AccordionGroup>
  <Accordion title="Marketing & Social Media">
    * Social media posts
    * Blog illustrations
    * Ad creatives
    * Thumbnail images
  </Accordion>

  <Accordion title="Design & Mockups">
    * Logo concepts
    * UI mockups
    * Product visualizations
    * Presentation graphics
  </Accordion>

  <Accordion title="Creative Projects">
    * Story illustrations
    * Concept art
    * Character designs
    * Album artwork
  </Accordion>

  <Accordion title="Productivity">
    * Presentation visuals
    * Document illustrations
    * Diagram concepts
    * Avatar generation
  </Accordion>
</AccordionGroup>

## Configuration

### Settings → Image Generation

| Setting           | Description                    |
| ----------------- | ------------------------------ |
| **Default Model** | Preferred generation model     |
| **Default Size**  | Standard image dimensions      |
| **Quality**       | Default quality setting        |
| **Save Location** | Where to save generated images |
| **Auto-save**     | Automatically save all images  |

## Best Practices

<AccordionGroup>
  <Accordion title="Be descriptive">
    More detail usually produces better results. Describe colors, lighting, mood, and style.
  </Accordion>

  <Accordion title="Use references">
    "In the style of \[artist/style]" can guide the aesthetic.
  </Accordion>

  <Accordion title="Iterate">
    Start with a basic prompt and refine based on results.
  </Accordion>

  <Accordion title="Avoid prohibited content">
    AI image generators have content policies. Avoid generating harmful or inappropriate content.
  </Accordion>
</AccordionGroup>

## Limitations

* Cannot generate photorealistic faces of real people
* Text in images may be inaccurate
* Complex scenes may have artifacts
* Generation time varies by model and quality

## Credits & Pricing

| Model            | Credit Cost       |
| ---------------- | ----------------- |
| DALL-E 3         | \~50 points/image |
| DALL-E 2         | \~20 points/image |
| Stable Diffusion | \~10 points/image |

<Note>
  Point costs vary by resolution and quality settings. Check Settings → Usage for your current balance.
</Note>

## Related Features

<CardGroup cols={2}>
  <Card title="OCR" icon="text" href="/ai/ocr">
    Extract text from images
  </Card>

  <Card title="SmartBar" icon="magnifying-glass" href="/features/smartbar">
    Quick image generation
  </Card>
</CardGroup>