# Prompt Guide - Realistic Mode

Prompts allow you to guide the AI in reimagining or modifying your image using natural language.\
A well-written prompt makes the difference between a generic result and a coherent, high-quality output.

{% columns %}
{% column width="50%" %}

<figure><img src="/files/fljWfn1cY3VvBEJsZjGZ" alt=""><figcaption></figcaption></figure>
{% endcolumn %}

{% column width="50%" %}

#### How Prompts and Creativity Work Together

The **creativity setting** directly affects how much influence your prompt has on the final image.

* With **higher creativity**, the AI interprets your prompt more freely — it can imagine new elements, adjust composition, or add details that weren’t in the original.
* With **lower creativity**, the AI focuses on fidelity — your prompt still applies, but only subtly, keeping the result closer to the source image.

In short, *the more creativity you allow, the stronger your prompt’s impact will be.*
{% endcolumn %}
{% endcolumns %}

{% hint style="success" %}
In short, *the more creativity you allow, the stronger your prompt’s impact will be.*
{% endhint %}

***

#### 1. Be Specific

Use clear, descriptive language.\
Include details such as colors, materials, style, and subject attributes.\
Avoid vague instructions like “make it better.” Instead, write something like:\
“enhance the texture of the leather jacket and increase lighting contrast.”

**Example:**\
“portrait of a woman with short black hair wearing a red coat, cinematic lighting”

***

#### 2. Preserve What Matters

Explicitly mention what should stay unchanged.\
This helps the AI maintain key elements such as composition, facial identity, or structure.

**Example:**\
“change the background to a beach while keeping the person’s pose and facial features identical”

***

#### 3. Edit Text or Objects Precisely

When replacing text or modifying details, quote the exact change:\
“replace ‘old text’ with ‘new text’.”\
Keep fonts readable and text length similar to preserve the layout.

For object edits, describe the action clearly:\
“add sunglasses to the man’s face” or “change the jacket color to dark blue.”

***

#### 4. Style and Artistic Control

Describe the artistic direction using recognizable visual terms or art movements:\
“oil painting, visible brushstrokes, baroque lighting” or “1960s pop-art poster.”\
Avoid generic words like “artistic” — specify the exact look or atmosphere you want.

***

#### 5. Handle Complex Edits Step by Step

When your edit involves multiple visual changes (background, lighting, or clothing), separate them into smaller actions.\
Start simple, confirm the result, then refine it.\
This approach gives the AI clearer guidance and more consistent results.

***

#### 6. Use Action-Based Language

Prefer **verbs** that describe specific actions: “add,” “replace,” “adjust,” “enhance,” “remove.”\
Avoid vague terms like “transform.”\
Precise instructions make the model’s behavior more predictable.

***

#### 7. Quick Tips Summary

* Be detailed and specific about **colors, styles, and subjects.**
* **Creativity and prompt strength go hand in hand:** higher creativity gives the prompt more influence, lower creativity keeps it faithful.
* Clearly state what should stay **unchanged.**
* Use quotation marks for **text replacements.**
* Mention **lighting or camera angle** when relevant.
* Work **progressively** — simple edits first, then refine.


---

# 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://lupaupscaler.gitbook.io/lupaupscaler-docs/prompt-guide-realistic-mode.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.
