Sexy AI User Guide

Getting Started

1.1. Introduction to Image Generation

At Sexy AI, you can use many different options to get the output you desire. Let's start with a quick overview of various elements you'll encounter.

Workflows

Workflow refers to the core AI engine (e.g., Sexy 1.0, Pony, SDXL). Each workflow has its own set of training data and is unique from other workflows in what and how it will perform. Depending on what you are trying to accomplish, one workflow may be more effective than another.

Models

Model refers to a style preset within a workflow (Dreamshaper, Lustify, PonyRealism, etc.). Model availability depends on its compatibility with the workflow (core AI engine) selected.

Prompts

Prompt refers to the text input used to influence the generation. A positive prompt tells the AI what you do want to see, and a negative prompt indicates what you do not want to see in the final output.

Settings

Setting refers to a modifier that can be applied to a generation (e.g., batch size, clip skip, CFG scale, LoRAs).

1.2. Platform Tour & Account Setup

Membership Tier Overview and Subscribing

Signing in without subscribing allows you to browse and react in Communities (emoji reactions). Tier 1 and Tier 2 members can generate, view libraries, and post/comment/create in Communities.

Generally, Tier 2 provides members with double the GPU time per rolling 24-hour period, priority queuing, and double concurrent generations compared to Tier 1. For up-to-date information on membership tiers and their features, please visit Account > Billing.

Subscriptions

Currently, there are two payment processors available for subscribing:

• Epoch allows you to set up recurring payments that will be debited from your account on a monthly or annual basis depending on your preference.
• G2A allows you to purchase a gift card/key that can be redeemed for a one-time purchase (1-month, 3-month, or 1-year).

Subscriptions FAQs

I can't use Epoch or G2A. What can I do?
We only accept payment through Epoch and G2A. If you are encountering an error when trying to purchase a subscription, you can reach out to support@sexy.ai or to the payment processor directly for assistance.

If I cancel, will it deactivate my membership?
You will be able to continue using Sexy AI until the end of your paid billing cycle.

How do I upgrade from Tier 1 to Tier 2 or downgrade from Tier 2 to Tier 1?
For either upgrading or downgrading, you'll need to cancel your current active subscription by logging into your Epoch dashboard. You can then purchase a new subscription once your current paid period ends (expiration date located under Account > Billing).
If you'd like to upgrade prior to the expiration date, please reach out to support@sexy.ai and indicate that you have canceled and are ready to upgrade.

Dashboard Navigation

Workflow Selector

In the header bar (top of page), there is a dropdown menu that allows you to select the workflow you'd like to use. To select a workflow:

1. Click the dropdown menu
2. Click the workflow you'd like to use

Model Dropdown

For most workflows, you'll see a dropdown menu at the top of the box beneath the header bar. To select a model:

1. Click the dropdown menu
2. Click the model you'd like to use

Settings Panel

The options in the settings panel will change depending on the workflow you select. Most workflows include a positive and negative prompt section. Here is a list and brief description of other settings you might see:

• Height and Width: Dimensions of the content you'd like to generate
• Batch Size: The number of images you'd like to make in one generation (creating a higher number of images does not impact the GPU time used - it is more efficient to select a higher number)
• CFG: How closely you would like the AI to follow your prompt (higher is not always better)
• Seed, Subseed, Seed Strength: Control randomness and reproducibility
• Clip Skip: Controls how literally the AI interprets your prompt
• Sampler: Algorithm used for image generation
• Destination: Where you want your content to go in your Library/Folders once generated

Settings Deep Dive

CFG Scale

CFG Scale (short for Classifier-Free Guidance) controls how closely the AI follows your prompt when generating an image.

Low CFG (1–4) → The AI takes more creative freedom. Results might be more abstract, loose, or unexpected — but also sometimes less accurate or relevant.

Medium CFG (5–8) → A balanced result. The AI generally follows your prompt while still adding some creative interpretation.

High CFG (9–12) → The AI follows your prompt very closely. Images tend to be sharper, more vivid, and more literal — but can become overly intense or rigid. In some cases, CFG set too high can result in a 'bad' generation (blurs, static).

Most users find a sweet spot between 7 and 9, but it's worth experimenting depending on your prompt.

CLIP Skip

CLIP Skip controls how literally the AI interprets your prompt when generating images. It adjusts how many layers of your prompt the AI "reads" or processes, which can influence how tightly or loosely it sticks to what you wrote.

LoRA

LoRA stands for Low-Rank Adaptation, a method used to enhance AI-generated images by adding specialized, lightweight training data to a base model. Think of a LoRA as a focused style or concept add-on — it helps the AI better understand and render specific characters, poses, styles, or features that the base model alone may not handle well.

• LoRAs can be selected from a dropdown list or added manually using a hash code (found on sites like Civitai) depending on the workflow/model.
• You typically control how strong the LoRA's influence is using a strength value (e.g., 0.1 to 2.0).
• Some LoRAs may require trigger words in your prompt to activate their full effect.

Please be sure that the LoRA you choose follows Sexy AI's Terms of Service to prevent any negative moderation action such as content removal or suspension.

Steps

Steps control how many times the AI refines an image before finishing it. Each step is like a brushstroke in a multi-layered drawing — the more steps, the more detail the AI can add.

Pro Tip: More steps don't always mean a better image — sometimes they over-refine the image, making it look too soft or unnatural. Test what works best for your prompt!

Sexy 1.0 Workflow Overview

Good general-purpose engine that balances quality and speed.

Good For:

• Photorealistic scenes, illustrations, fantasy art
• Quickly iterating on concepts, styles, and compositions

Limitations:

• Good hands & faces
• Text in images
• Complex scenes - Multi‑character compositions or intricate interactions may blur or overlap unintentionally
• Non-square outputs
• Over‑saturation & artifacts - High CFG or extreme prompts can lead to oversharpened edges, noise, or color clipping

Prompting Tips for Sexy 1.0

Follow a clear descriptor order

Structure your prompt from broad to specific. A common pattern is:

Subject → Context/Action → Style/Artist → Lighting/Color → Camera/Composition → Details

Example: "A serene forest clearing with a woman standing in mist (subject/context), oil painting style, John William Waterhouse style (style/artist), soft golden morning light, 35 mm photograph composition, detailed bark and foliage textures."

Use comma or pipe separators

• Commas (… , …) for natural breaks
• Pipes (… | …) when you want to group tightly related concepts

Example: "Cyberpunk cityscape, neon reflections, rainy night | (blade runner style:1.2)"

Leverage weight modifiers

• Parentheses increase emphasis: (element:1.3)
• Brackets decrease emphasis: [element:0.7]

Example: "(dramatic lighting:1.4), [background clutter:0.6]"

Additional Best Practices:

• Keep CFG around 7–9 for balanced creativity vs. fidelity
• Use steps in the 25–35 range to avoid under‑ or over‑refinement
• Include "masterpiece" or "best quality" judiciously as generic quality boosters
• Watch prompt length - aim for 3–7 well‑chosen phrases rather than a single run‑on sentence
• Experiment and iterate - small tweaks can yield surprisingly different results

Prompting 101

To create good images with AI, you will need to understand how to properly craft your prompt. This means you must understand the following concepts:

• The components of a positive prompt
• The usage of negative prompts
• Special operators and image generation steps

Components of a Positive Prompt

Your positive prompt should be a short and concise list of terms that describe the elements you want to see in your image. You should start with the following types of terms:

Positive Prompt Examples

• a painting of an elf, realistic
• a realistic photo of an elf
• an abstract drawing of an elf

Positive Prompt Examples

• a photo of an elf,sitting on the ground,forest background,realistic,detailed
• a photograph of an exhausted elf, siting in a dark forest,hyperrealistic,detailed
• an abstract forest background, an elf,gorgeous face,alluring features,ultradetailed,ultraquality

Components of a Negative Prompt

Because of the wide range of concepts associated with certain terms, you may see elements you don't want in your final image. Negative prompts can be helpful in clarifying what you don't want to see in your final image.

Negative Prompt Example

Elves are both creatures that exist in high-fantasy settings but also as part of Christmas mythology. The AI may therefore combine elements of both into your image. However, you can add Christmas to the negative prompts to get a result like the following image

Positive an elf, in a cottage

Negative Christmas

Negative Prompt Tips

When it comes to using negative prompts, you should keep a few things in mind

1. Keep the negative prompt list short.
2. Always start an image with no more than three negative prompts and only add to the list as needed.
3. Avoid conflicts between your positive and negative prompts.
4. Be as broad as possible with your negative prompts.

Keyword Conflict

1. Two or more keywords that produce image elements that contradict each other.
2. Without proper weighting and balance the contradicting terms will distort the image, and produce unwanted results.
3. It is advised, when creating a new prompt to take the time and add each keyword at a time to check for problem terms.

Keyword conflict example

Special Syntax

Special syntax refer to certain structures and punctuation in your prompts that the AI interprets in a specific way. The most basic special operator is a comma ,, telling the AI where one part of your prompt ends and the next part begins. However, there are a variety of other special operators you can use to add nuance to your prompts and return a better image.

Avoid using more than 3 layers of brackets or parentheses as this may create unexpected results. Additionally, weights may behave differently when used in negative prmopts compared to positive prompts. Test different weights to experiment with the result.

Keyword Weight Prompt Examples

Let's start with the basic prompt,

a woman, garden background

Now let's tone down the weight of a woman.

[a woman], garden background

• Notice how the garden background is now more apparent in the image.*

Now let's tone down the weight of garden background

a woman, [garden background]

Notice the woman being more in focus than the garden.

Additional Syntax

While the usage of parentheses and brackets is universal, some models support additional syntax. You can experiment with the following syntax:

• (term:X.X), where X.X is a decimal number: This multiplies the weight of your term by the value of X.X. A value less than one decreases the weight and a value greater than one increases the weight. It is recommended to keep X.X between 0.5 and 1.5

Prompt Blending

Prompt blending takes two or more concepts and merges them together in the final image. For example, you could use prompt blending syntax with apple and fire to create an apple that appears to be on fire.

[term1:term2:X.X]

• Combines term1 and term2 with blending control via X.X.
• When X.X ranges from 0.2 to 0.5, term2 is favored.
• For values between 0.5 and 0.9, term1 is favored.

[keyword1:keyword2:0.3]

• Emphasizes keyword2 more.

[keyword1:keyword2:0.7]

• Emphasizes keyword1 more.

[term1|term2|term3]

• Blends three terms equally.

{term1|term2|term3}

• Requires the terms to appear/happen together.

{term1:term2}

• Blends the terms with equal weight.

(credit Meaca_g)

Steps and Weights

To understand the technical application of the syntax above, it's important to understand the basics of how the image is generated. At its core, the Stable Diffusion engine works on a series of steps. After each step, the image becomes more refined with diminishing returns as the number of steps increases. Sexy.ai generates images with 20 steps, meaning that your image is generated after 20 iterations.

Practically speaking, the special syntax modifies how terms are applied during the step process.

• A [term] will only be active for 0.9 of the frames, or 18 frames
• A (term) will be more active in all frames.
• A (term)[term2] will be averaged out.
• [term1|term2] will alternate every other step.
• [term:term2:X.X] will swap from term to term2 after the proportion given. For example, term1:term2:0.4 would swap from term1 to term2 after the 8th step (0.4 x 20 = 8)

Multiple Subject

Multiple subjects can be created by the following keywords and prompt phrases

Subject Combinations

Image Composition Tips

• The success of the image will depend on the type of shot, the engine, and ultimately the term usage.
• If you have an extreme close shot or close shot and are not getting two subjects, consider pulling back and trying a more open shot, such as long shot, wide shot, or far shot.
• Try not to be overly specific when requesting two people or more. More people means more pixel division.

Point of View/Perspective Shot

• POV/Perspective shots use a fisheye style lense to create a realistic voyueristic effect.
• POV shots will take the term two people/men/women etc as the viewer and the subject and may keep a third person from being added.

POV Basic Prompt Structure

Simple

• pov,[angle of view],view of [subject statement],[major/minor details], [setting],[lighting],[quality]

Advanced

• [gender][camera distance]pov,[angle of view],view of [subject statement],[major/minor details], [setting],[lighting],[quality]

Prompt Example

pov,from overhead,view of a woman laying on a bed,hotel lobby background, soft light, soft shadows, ultraquality >

Perspective Shots