# User Manual

Basic Commands

### `/image`

Generates an image based on the provided prompt text. It takes approximately 0.5-1 minute to produce an image.

### `/help`

Displays helpful information about Maze Guru.

<figure><img src="/files/ErT93Qtt2dmvjFJ1avN6" alt=""><figcaption></figcaption></figure>

### `/Settings`

Allows you to adjust your settings:

* Model: Select the painting model.
* Size: Select the image size.
* Quality: Select the image quality.
* Similarity: Only used in image-to-image generation. A higher value makes the generated image closer to the original, while a lower value makes it more creatively interpreted.

<figure><img src="/files/l3zyLEWwIUWaqKgqPH2A" alt=""><figcaption></figcaption></figure>

### `/info`

View the current account's painting usage:

* Current Plan: the subscription plans
* Status: fast mode or unlimited mode
* Current Available Points
* Subscription Points
* Rewarded Points

<figure><img src="/files/tLf6ClBi2UEAgImVh9YA" alt=""><figcaption></figcaption></figure>

### `/show`

Displays all the images you have generated in the MazeAI Bot DM.

<figure><img src="/files/3UB52R8ko749mimSyaG0" alt=""><figcaption></figcaption></figure>

Why am I not receiving messages in my private channel?&#x20;

Please check your Privacy & Safety settings and ensure that you have enabled "Allow direct messages from server members."

<figure><img src="/files/hY2oAaXlMZhGgcPFdyLD" alt=""><figcaption></figcaption></figure>

### **`/share`**

Shares your image in the showcase channel.

### **`/add-server`**

Adds the Maze Guru bot to your own server.

<figure><img src="/files/tEQslfOvFcYoNJTeylJE" alt=""><figcaption></figcaption></figure>

## Image parameter Illustrated

### Base Structure

Parameters that change how the images are generated. A complete /image command may contain various elements, such as an image URL, image weights, text prompt, style, and other switches.&#x20;

/image parameters should follow the order mentioned below.

<figure><img src="/files/4BfOfC7RdNlHCOfFv24W" alt=""><figcaption></figcaption></figure>

### **Seed**

The random seed determines the initialize noise pattern and hence the final image.

Setting it to -1 means using a random one every time. It is useful when you want to generate new images. On the other hand, fixing it would result in the same images in each new generation.

Use -`-seed`\<value> to the end of your prompt. for example:

`/image 1 girl --seed 123`

### **Negative prompt**

The negative prompt feature can help you easily remove unwanted content.

Use `--no`to the end of your prompt. for example:

<figure><img src="/files/exgFZZuqP2HZBQZ95fHp" alt=""><figcaption></figcaption></figure>

### Character Consistence

Character consistence feature allows you to create images that maintain consistency across characters

Simply add `--cref URL --cw 1-2` at the end of your prompt.

* `--cw 1`: Maintains facial consistency.
* `--cw 2`: Maintains consistency across the entire body.

### Stylization

use `--stylize` to adjust stylization

Higer the stylization is, more artistic the image will be.

Lower the stylization is, more closely the image will match the prompt.

### Style Random

With Style Random, you can now generate artwork without specifying an artistic style or artist in your prompt.

use `--sref random` at the end of your prompt

**This feature is only available for Midjourney and Niji models.**

### Create image with URL

Add one or more image URLs to your prompt, and the generated image will be visually inspired by those images. It's important to mix words with images.

<figure><img src="/files/VRhWDLPdRLIbcdZNriwa" alt=""><figcaption></figcaption></figure>

### Adjust image size

Maze Guru supports multiple size ratios:

* 1:1 (avatar)&#x20;
* 1:2 (phone)
* 3:4 (Ilustration)&#x20;
* 4:3 (Media)
* 9:16 (Poster)
* 16:9 (wallpaper)

Use the `--size`parameter when using `/image`command

<figure><img src="/files/6MkGb1taptvjLeHbEv5b" alt=""><figcaption></figcaption></figure>

### Style selection

To explore different image styles, you can set the `--style` parameter and experiment with the following supported styles:

| Mode   | Style                                                                                                                                                                                                                                  |
| ------ | -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| Stable | <p>Scifi</p><p>Fantasy</p><p>Real Photo</p><p>Asian Portrait</p><p>Girl Realistic</p><p>Portrait Album</p><p>Paper-cutting</p><p>Dreamwork </p><p>Mecha </p><p>Avatar</p><p>Illustration </p><p>Midjourney</p><p>Disco</p><p>SD2.0</p> |
| Anime  | <p>Semi Impasto</p><p>Digital Humans</p><p>Chinoiserie 2.5D</p><p>Flat CG Artwork</p><p>CG Artwork</p><p>2.2D</p><p>Indie Pop</p><p>Illustration</p><p>Anime plus </p><p>Anime 2.0</p><p>2.5D</p><p>Photo-Comic</p>                    |

## **What is Style Extension**

Style Extension is an amazing feature that allows you to create new styles and variations on the original model, resulting in stunning visual images with just ONE CLICK!

<figure><img src="/files/fr1fsMPlHOLnvMjWXi4Q" alt=""><figcaption></figcaption></figure>

### **How to use Style Extension**

**1. Style Extension Button**

Under certain styles, a new button will appear in the bottom right corner of the image after creation. Clicking on this button will prompt you to select a Style Extension option. Once you have chosen the desired type of style transfer extension, the system will generate new images based on your original settings.

<figure><img src="/files/Pf4tcWOUz47Kep6N3Cbx" alt=""><figcaption></figcaption></figure>

<figure><img src="/files/EmQIn6QpAkASFfyeP3Pz" alt=""><figcaption></figcaption></figure>

**2. Trigger word in the prompt**

Each Style Extension comes with its trigger keywords. Simply input the corresponding trigger keyword in your prompt, and the Style Extension functionality will automatically activate for your current creation. Note:&#x20;

1. Trigger keyword only works under the corresponding model.&#x20;
2. Any single trigger word can take effect.&#x20;
3. Trigger words need to be separated in the prompt.

<figure><img src="/files/1iCJ4gJsyTwEI7pdLj5e" alt=""><figcaption></figcaption></figure>

### **Avaliable Style Extension**

<table><thead><tr><th width="138">Style</th><th>Style Extension</th><th>Trigger Keyword</th><th>Scope of Application</th></tr></thead><tbody><tr><td>Fantasy</td><td>Impasto</td><td>Impasto</td><td>Anything</td></tr><tr><td>Fantasy</td><td>Mecha</td><td>mecha / bodysuit</td><td>Mecha and science fiction related</td></tr><tr><td>Fantasy</td><td>Strange things</td><td>a fantz creature wlop</td><td>Dark and mysterious things</td></tr><tr><td>Fantasy</td><td>Chibi doll</td><td>chibi wlop</td><td>Chibi people and things</td></tr><tr><td>Fantasy</td><td>Medieval</td><td>fantz monster / by mooncryptowow</td><td>Fantasy and medieval retro related</td></tr></tbody></table>


---

# Agent Instructions: Querying This Documentation

If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://docs.maze.guru/maze-guru/user-manual.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.