> For the complete documentation index, see [llms.txt](https://lupa-ai-documentation.gitbook.io/lupa-ai-documentation/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://lupa-ai-documentation.gitbook.io/lupa-ai-documentation/flux.md).

# Flux

<details>

<summary><strong>Overview</strong></summary>

Flux Mode is the most **balanced and versatile** option in Lupa AI.\
It is designed to perform well across nearly any image type — from renders and artwork to photos and marketing visuals — without introducing errors or distortion.

This mode is ideal when you want high quality output with minimal risk of:

* Artifacts
* AI hallucinations
* Layout distortions
* Over-enhancement

It combines the clarity of High Fidelity with some creative flexibility of the Creative mode, but always staying stable and professional.

**Recommended if you're unsure which mode to use.**

**New:** Portrait Sharpen

\
Flux now includes **Portrait Sharpen**, a specialized enhancement designed to improve **sharpness, skin texture and facial detail** in close-up images.\
It works especially well for:

* Studio-style portraits
* Fashion or product shots featuring faces
* Images where realism and sharpness are critical

⚠️ While it enhances skin and eyes with precision, Portrait Sharpen avoids artificial textures or stylistic effects — making it ideal for commercial and editorial use.

>

</details>

<details>

<summary><strong>Key Adjustable Settings</strong></summary>

* **Prompt (optional)**\
  A short description of what's in the image.\
  This helps guide the model when interpreting complex content.\
  Example: *“glass perfume bottle on a white background”*
  * *Detect from image =* AI automatically describes the selected image

<div align="left"><figure><img src="/files/DIYakC3e5E4iQ1UDb4FV" alt="" width="232"><figcaption></figcaption></figure> <figure><img src="/files/R4OVQD5uSJyOBAWr7LRU" alt="" width="349"><figcaption></figcaption></figure></div>

* **Scale Factor**\
  Choose between x2, x4, x6, x8 or x10 enlargement.\
  Larger scale increases resolution and detail — but also processing time and credit usage.

<div align="left"><figure><img src="/files/M95fhMHpIPycspeGnFux" alt=""><figcaption></figcaption></figure></div>

* **Creativity (slider)**\
  This controls how much freedom the AI has to modify details:\
  \- Low = faithful upscaling with subtle fixes\
  \- High = deeper transformation and enhancement\
  ⚠️ Use with caution if you want to preserve the original image closely.

<div align="left"><figure><img src="/files/FFJROlUTFR6ysKeV7X1z" alt=""><figcaption></figcaption></figure></div>

#### Portrait Sharpen (toggle)

This is a dedicated enhancement toggle for portraits and close-up faces.\
When enabled, it improves **sharpness**, **skin texture**, and **fine facial detail**, avoiding over-stylization.

To activate it:

> Click the **"Portrait Sharpen"** button that appears after selecting **Flux** mode.

<figure><img src="/files/MgoKrlSAE6VO05B3wtqv" alt=""><figcaption></figcaption></figure>

Creative mode:

* Offers the most artistic freedom.
* The AI may slightly reinterpret facial features, textures, and lighting.
* Best for stylized portraits or fantasy results.
* **Not recommended** for blurry or distant faces, as it may generate unwanted artifacts.

Ultra Leal Mode:

* Prioritizes **faithfulness** to the original face.
* Keeps proportions, structure, and features as close to the source as possible.
* Very low tolerance for hallucination.
* Also **not ideal for blurry or low-res faces**, as it tries to preserve what’s already there without much invention.

Ultra Real Mode:&#x20;

* Strikes a balance between realism and enhancement.
* Adds realistic **4K skin texture**, refines facial features, and enhances eyes and hair.
* May **slightly alter facial features** for a more photographic or commercial look.
* Ideal for **improving AI-generated or soft-looking faces** with added realism.

</details>

<details>

<summary><strong>Best Use Cases</strong></summary>

* Concept art or environment scenes

<div align="left"><figure><img src="/files/zGZ9R11bGjOIbsJ0Snte" alt=""><figcaption></figcaption></figure></div>

<figure><img src="/files/qTsrmMaXAWsBU3DV4DZ2" alt=""><figcaption></figcaption></figure>

<figure><img src="/files/AddITpXP1t8RjB1aI4pP" alt=""><figcaption></figcaption></figure>

<figure><img src="/files/UqCMJsDBikWVqRji2Xx1" alt=""><figcaption></figcaption></figure>

<figure><img src="/files/4V7oayH9SQGjO2by02ZA" alt=""><figcaption></figcaption></figure>

<figure><img src="/files/r16ACB5mN9cvJARMDWWM" alt=""><figcaption></figcaption></figure>

* Photos with moderate quality issues

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

<figure><img src="/files/fVMr0DYmo7rOJN9QnOFK" alt=""><figcaption></figcaption></figure>

<figure><img src="/files/OoWeKlJv33Hb8ktLVRhE" alt=""><figcaption></figcaption></figure>

* Portrait (Portrait Sharpen Activated)

<figure><img src="/files/DrTzR9giOjH48QpibWjs" alt=""><figcaption></figcaption></figure>

<figure><img src="/files/EUIjTeYvDpmh0JaGb34w" alt=""><figcaption></figcaption></figure>

<figure><img src="/files/PfTELwsNTS6XQYCpT0Wn" alt=""><figcaption></figcaption></figure>

<figure><img src="/files/6q4aJakdhgvZqTbgcdY6" alt=""><figcaption></figcaption></figure>

⚠️**Not recommended for:**

* Stylized artwork needing texture addition → use Creative
* Extreme abstraction or surreal edits

</details>


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## 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://lupa-ai-documentation.gitbook.io/lupa-ai-documentation/flux.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.