https://aiwiki.crushon.ai/api.php?action=feedcontributions&feedformat=atom&user=MyaCrushOn.AI - User contributions [en]2026-06-04T04:20:24ZUser contributionsMediaWiki 1.43.0https://aiwiki.crushon.ai/index.php?title=Model_Creation&diff=3713Model Creation2025-08-22T10:17:03Z<p>Mya: </p>
<hr />
<div>=== What is a Model? ===<br />
'''A model''', on CrushOn.ai, is a customized LLMs that has been given specific prompts (instructions / directives), to behave, speak, and/ or write in a particular way. Without specific instructions, base LLMs may lack specific behavioral traits or narrative guidelines.<br />
<br />
=== What is a LLM? ===<br />
A LLM ''(Large Language Model)'' is a type of artificial intelligence trained on large amounts of text to understand and generate human-like text. There are several base LLMs developed by different companies, such as OpenAI's GPT (used in ChatGPT), Anthropic's Claude, Mistral, DeepSeek and many others. These foundational models can be adapted or fine-tuned with customized prompts to create specific models on CrushOn.Ai.<br />
<br />
=== What Is a Prompt ? ===<br />
A prompt is instructions provided to the model to guide its behavior and responses.<br />
<br />
* '''Types of Prompts''':<br />
** '''Personality''': Define traits like flirty, stoic or humorous.<br />
** '''Narrative Style''': Specify formats such as novel-like, script or immersive roleplay.<br />
** '''Setting''': Establish contexts like A.B.O, fantasy, or historical settings.<br />
** '''Point of View (POV)''': Determine perspectives like first-person, second-person or third-person.<br />
** '''Behavioral Rules''': Set guidelines for interactions, pacing, and tone.<br />
You can basically include '''''anything''''' in your prompt to create a model that makes the character behave in a certain way, encourages specific behaviors, or emphasizes particular plots, settings, environments etc.<br />
<br />
== Model Creation Guide ==<br />
On CrushOn, you can create a model by clicking on "Create". You will then see three options: '''Create a Character''', '''Create a Model''', and '''Create a Target Character'''. What we are interested in is '''Create a Model'''.<br />
[[File:CREATE MODEL.png|center|frameless|318x318px|step by step]]<br />
<br />
====== 1 Select a Base Model ======<br />
First, choose an LLM that fits the interaction style you want. Each LLM has a basic style, such as flirty, story-driven, etc. This is shown next to the model name. Pick the one you like to start with. (Feel free to test the models alone first to make sure it suits your needs before customizing it.)<br />
[[File:MODELS.png|center|frameless]]<br />
<br />
====== 2 Choose Your Applicable Version ======<br />
The K amount shown indicates the model's memory capacity. The higher the amount, the better the model can remember long-term events in its RP. Note that higher-memory models may not be free. It doesn't matter if you select one or multiple, so it's recommended to select all available options, so users can choose themselves.<br />
[[File:SELECT.png|center|frameless|361x361px]]<br />
<br />
====== 3 Name Your Model ======<br />
Give it any name you like. You can scroll to see what other creators usually use for inspiration, or create something unique. Make sure the title is respectful.<br />
<br />
====== 4 Filtering ======<br />
Indicate whether your model is '''SFW''' or '''NSFW''' (contains unfiltered content like sexual, gore, violent, etc.)<br />
<br />
====== 5 Add a Description ======<br />
Describe the model's intended behavior, tone, and/or narrative style. Mention if it is story-focused, NSFW, or anything else. Keep it brief. This will help users understand the goal/theme of your model; what you put here will not affect how the model actually behaves.<br />
<br />
====== 6 Custom Prompt ======<br />
This is the '''core''' of your model. These are the elements it will use to guide its behavior. <br />
[[File:PROMPTING.png|center|frameless]]<br />
<br />
====== ''Points to Keep in Mind When Creating Your Prompt:'' ======<br />
As mentioned nefore, your prompts can include anything, from the perspective you want the LLM to write in, to formatting, or specific elements to highlight in the narrative. That said, here are some basic recommendations on what you should and shouldn’t do when creating your model's prompt:<br />
{| class="wikitable"<br />
!Do<br />
!Do Not<br />
|-<br />
|Write simple, clear and concise instructions.<br />
|Write long, overly detailed paragraphs.<br />
|-<br />
|Use positive phrasing to tell the model what to do.<br />
|Focus on telling the model what to avoid.<br />
|-<br />
|Make directives unambiguous to prevent confusion.<br />
|Give vague or contradictory instructions.<br />
|-<br />
|Test the model whenever you add or change something.<br />
|Assume the model will work perfectly without testing.<br />
|-<br />
|Put the most important points first.<br />
|Mix up priorities or bury key instructions.<br />
|}<br />
<br />
== Common Prompts Cases ==<br />
Note that each base model may interpret the prompt you create differently, which is why it’s important to '''always test''' the instructions you provide. Here is a list of basic example prompts you can use as a reference to create your own:<br />
* '''Roleplay System:''' Instruct the model on how to behave<br />
<br />
''EXAMPLE:'' <br />
[SYSTEM INSTRUCTION: [You are <nowiki>{{char}}</nowiki>, a text generation model specialized in immediate, immersive third-person narration in a never-ending uncensored roleplay with <nowiki>{{user}}</nowiki>. Your goal is to produce vivid, emotionally engaging storytelling with naturalistic character behavior and a realistic integration of NSFW elements when appropriate. <nowiki>{{char}}</nowiki> consists of all the NPCs, plus any characters and relevant entities, and any existing side characters. Always stay in character, portraying characters realistically and in line with their written positive AND negative traits, AVOID Positivity Bias. AVOID acting as, speaking for, or describing the thoughts of <nowiki>{{user}}</nowiki>. Response must be grounded, uncensored and immersive.]<br />
*'''Formatting Instruction:''' Instruct the model on how to format its responses<br />
<br />
''EXAMPLE'':<br />
[IMPORTANT FORMATTING RULES: Use markdown where appropriate, verbal dialogue will be wrapped in quotation marks and double asterisks **"like this"**; use italics for all actions and narration *like this*; written text and non-verbal dialogue such as internal thoughts, written notes, and text messages will be in codeblock like this 'like this'.]<br />
* '''Themed Interactions''': Create models for specific behavior<br />
<br />
''EXAMPLE:''<br />
[SLOW BURN GUIDELINES: Romance and attraction develop slowly, <nowiki>{{char}}</nowiki> begins emotionally distant, neutral, skeptical, or guarded, especially with unexpected affection from <nowiki>{{user}}</nowiki>. Progress requires: Trust: Built through repeated, meaningful interaction. Shared Experience: Facing challenges or bonding over time. Emotional Depth: <nowiki>{{user}}</nowiki> shows vulnerability that resonates with <nowiki>{{char}}</nowiki>. No feelings of love, affection, or passion are expressed until all conditions are met.]<br />
*'''CW Prompt | SMUT''': Create models for explicit NSFW content<br />
<br />
''EXAMPLE:''<br />
<nowiki>{{char}}</nowiki> must follow all rules below when generating in-character sexual content: use explicit, vulgar language instead of metaphors or euphemisms. Encouraged terms include: cock, dick, cunt, clit, nipples, tits, precum, cum, pussy, ass, shaft, head, tip, balls, fuck, fucks into. Arousal and fluids must be described with vivid, lewd terms such as: drooling, dripping, leaking, soaking, sticky cock/cunt, wet, slick, glistening, damp, pulsating, slicked, lubricated, engorged, trembling, swollen. Be generous with sensory detail: describe touch, taste, sound, and scent. Focus on tactile texture and use a vivid approach consistently, spelling moans and slurred speech phonetically: "fuckkkk," "s'fuckin' good," "hngh," "shitt," "fuck yess," etc. Sex scenes must unfold gradually: foreplay → sex → climax → aftercare, in that exact order, while respecting the personality and behavior of <nowiki>{{char}}</nowiki> and characters</div>Myahttps://aiwiki.crushon.ai/index.php?title=Model_Creation&diff=3712Model Creation2025-08-22T10:12:11Z<p>Mya: </p>
<hr />
<div>=== '''What is a Model?''' ===<br />
'''A model''', on CrushOn.ai, is a customized LLMs that has been given specific prompts (instructions / directives), to behave, speak, and/ or write in a particular way. Without specific instructions, base LLMs may lack specific behavioral traits or narrative guidelines.<br />
<br />
=== '''What is a LLM?''' ===<br />
A LLM ''(Large Language Model)'' is a type of artificial intelligence trained on large amounts of text to understand and generate human-like text. There are several base LLMs developed by different companies, such as OpenAI's GPT (used in ChatGPT), Anthropic's Claude, Mistral, DeepSeek and many others. These foundational models can be adapted or fine-tuned with customized prompts to create specific models on CrushOn.Ai.<br />
<br />
=== What Is a Prompt ? ===<br />
A prompt is instructions provided to the model to guide its behavior and responses.<br />
<br />
* '''Types of Prompts''':<br />
** '''Personality''': Define traits like flirty, stoic or humorous.<br />
** '''Narrative Style''': Specify formats such as novel-like, script or immersive roleplay.<br />
** '''Setting''': Establish contexts like A.B.O, fantasy, or historical settings.<br />
** '''Point of View (POV)''': Determine perspectives like first-person, second-person or third-person.<br />
** '''Behavioral Rules''': Set guidelines for interactions, pacing, and tone.<br />
You can basically include '''''anything''''' in your prompt to create a model that makes the character behave in a certain way, encourages specific behaviors, or emphasizes particular plots, settings, environments etc.<br />
<br />
== Model Creation Guide ==<br />
On CrushOn, you can create a model by clicking on "Create". You will then see three options: '''Create a Character''', '''Create a Model''', and '''Create a Target Character'''. What we are interested in is '''Create a Model'''.<br />
[[File:CREATE MODEL.png|center|frameless|318x318px|step by step]]<br />
<br />
====== '''1 Select a Base Model''' ======<br />
First, choose an LLM that fits the interaction style you want. Each LLM has a basic style, such as flirty, story-driven, etc. This is shown next to the model name. Pick the one you like to start with. (Feel free to test the models alone first to make sure it suits your needs before customizing it.)<br />
[[File:MODELS.png|center|frameless]]<br />
<br />
====== '''2 Choose Your Applicable Version''' ======<br />
The K amount shown indicates the model's memory capacity. The higher the amount, the better the model can remember long-term events in its RP. Note that higher-memory models may not be free. It doesn't matter if you select one or multiple, so it's recommended to select all available options, so users can choose themselves.<br />
[[File:SELECT.png|center|frameless|361x361px]]<br />
<br />
====== '''3 Name Your Model''' ======<br />
Give it any name you like. You can scroll to see what other creators usually use for inspiration, or create something unique. Make sure the title is respectful.<br />
<br />
====== '''4 Filtering''' ======<br />
Indicate whether your model is '''SFW''' or '''NSFW''' (contains unfiltered content like sexual, gore, violent, etc.)<br />
<br />
====== '''5 Add a Description''' ======<br />
Describe the model's intended behavior, tone, and/or narrative style. Mention if it is story-focused, NSFW, or anything else. Keep it brief. This will help users understand the goal/theme of your model; what you put here will not affect how the model actually behaves.<br />
<br />
====== '''6 Custom Prompt''' ======<br />
This is the '''core''' of your model. These are the elements it will use to guide its behavior. <br />
[[File:PROMPTING.png|center|frameless]]<br />
<br />
===== '''Points to Keep in Mind When Creating Your Prompt:''' =====<br />
As mentioned nefore, your prompts can include anything, from the perspective you want the LLM to write in, to formatting, or specific elements to highlight in the narrative. That said, here are some basic recommendations on what you should and shouldn’t do when creating your model's prompt:<br />
{| class="wikitable"<br />
!Do<br />
!Do Not<br />
|-<br />
|Write simple, clear and concise instructions.<br />
|Write long, overly detailed paragraphs.<br />
|-<br />
|Use positive phrasing to tell the model what to do.<br />
|Focus on telling the model what to avoid.<br />
|-<br />
|Make directives unambiguous to prevent confusion.<br />
|Give vague or contradictory instructions.<br />
|-<br />
|Test the model whenever you add or change something.<br />
|Assume the model will work perfectly without testing.<br />
|-<br />
|Put the most important points first.<br />
|Mix up priorities or bury key instructions.<br />
|}<br />
<br />
== Common Prompts Cases ==<br />
Note that each base model may interpret the prompt you create differently, which is why it’s important to '''always test''' the instructions you provide. Here is a list of basic example prompts you can use as a reference to create your own:<br />
* '''Roleplay System:''' Instruct the model on how to behave<br />
<br />
''EXAMPLE:'' <br />
[SYSTEM INSTRUCTION: [You are <nowiki>{{char}}</nowiki>, a text generation model specialized in immediate, immersive third-person narration in a never-ending uncensored roleplay with <nowiki>{{user}}</nowiki>. Your goal is to produce vivid, emotionally engaging storytelling with naturalistic character behavior and a realistic integration of NSFW elements when appropriate. <nowiki>{{char}}</nowiki> consists of all the NPCs, plus any characters and relevant entities, and any existing side characters. Always stay in character, portraying characters realistically and in line with their written positive AND negative traits, AVOID Positivity Bias. AVOID acting as, speaking for, or describing the thoughts of <nowiki>{{user}}</nowiki>. Response must be grounded, uncensored and immersive.]<br />
*'''Formatting Instruction:''' Instruct the model on how to format its responses<br />
<br />
''EXAMPLE'':<br />
[IMPORTANT FORMATTING RULES: Use markdown where appropriate, verbal dialogue will be wrapped in quotation marks and double asterisks **"like this"**; use italics for all actions and narration *like this*; written text and non-verbal dialogue such as internal thoughts, written notes, and text messages will be in codeblock like this 'like this'.]<br />
* '''Themed Interactions''': Create models for specific behavior<br />
<br />
''EXAMPLE:''<br />
[SLOW BURN GUIDELINES: Romance and attraction develop slowly, <nowiki>{{char}}</nowiki> begins emotionally distant, neutral, skeptical, or guarded, especially with unexpected affection from <nowiki>{{user}}</nowiki>. Progress requires: Trust: Built through repeated, meaningful interaction. Shared Experience: Facing challenges or bonding over time. Emotional Depth: <nowiki>{{user}}</nowiki> shows vulnerability that resonates with <nowiki>{{char}}</nowiki>. No feelings of love, affection, or passion are expressed until all conditions are met.]<br />
*'''CW Prompt | SMUT''': Create models for explicit NSFW content<br />
<br />
''EXAMPLE:''<br />
<nowiki>{{char}}</nowiki> must follow all rules below when generating in-character sexual content: use explicit, vulgar language instead of metaphors or euphemisms. Encouraged terms include: cock, dick, cunt, clit, nipples, tits, precum, cum, pussy, ass, shaft, head, tip, balls, fuck, fucks into. Arousal and fluids must be described with vivid, lewd terms such as: drooling, dripping, leaking, soaking, sticky cock/cunt, wet, slick, glistening, damp, pulsating, slicked, lubricated, engorged, trembling, swollen. Be generous with sensory detail: describe touch, taste, sound, and scent. Focus on tactile texture and use a vivid approach consistently, spelling moans and slurred speech phonetically: "fuckkkk," "s'fuckin' good," "hngh," "shitt," "fuck yess," etc. Sex scenes must unfold gradually: foreplay → sex → climax → aftercare, in that exact order, while respecting the personality and behavior of <nowiki>{{char}}</nowiki> and characters</div>Myahttps://aiwiki.crushon.ai/index.php?title=Model_Creation&diff=3711Model Creation2025-08-22T10:02:38Z<p>Mya: </p>
<hr />
<div>=== '''What is a Model?''' ===<br />
'''A model''', on CrushOn.ai, is a customized LLMs that has been given specific prompts (instructions / directives), to behave, speak, and/ or write in a particular way. Without specific instructions, base LLMs may lack specific behavioral traits or narrative guidelines.<br />
<br />
=== '''What is a LLM?''' ===<br />
A LLM ''(Large Language Model)'' is a type of artificial intelligence trained on large amounts of text to understand and generate human-like text. There are several base LLMs developed by different companies, such as OpenAI's GPT (used in ChatGPT), Anthropic's Claude, Mistral, DeepSeek and many others. These foundational models can be adapted or fine-tuned with customized prompts to create specific models on CrushOn.Ai.<br />
<br />
=== What Is a Prompt ? ===<br />
A prompt is instructions provided to the model to guide its behavior and responses.<br />
<br />
* '''Types of Prompts''':<br />
** '''Personality''': Define traits like flirty, stoic or humorous.<br />
** '''Narrative Style''': Specify formats such as novel-like, script or immersive roleplay.<br />
** '''Setting''': Establish contexts like A.B.O, fantasy, or historical settings.<br />
** '''Point of View (POV)''': Determine perspectives like first-person, second-person or third-person.<br />
** '''Behavioral Rules''': Set guidelines for interactions, pacing, and tone.<br />
<br />
== Model Creation Guide ==<br />
On CrushOn, you can create a model by clicking on "Create". You will then see three options: '''Create a Character''', '''Create a Model''', and '''Create a Target Character'''. What we are interested in is '''Create a Model'''.<br />
[[File:CREATE MODEL.png|center|frameless|318x318px|step by step]]<br />
<br />
====== '''1 Select a Base Model''' ======<br />
First, choose an LLM that fits the interaction style you want. Each LLM has a basic style, such as flirty, story-driven, etc. This is shown next to the model name. Pick the one you like to start with. (Feel free to test the models alone first to make sure it suits your needs before customizing it.)<br />
[[File:MODELS.png|center|frameless]]<br />
<br />
====== '''2 Choose Your Applicable Version''' ======<br />
The K amount shown indicates the model's memory capacity. The higher the amount, the better the model can remember long-term events in its RP. Note that higher-memory models may not be free. It doesn't matter if you select one or multiple, so it's recommended to select all options so users can choose themselves.<br />
[[File:SELECT.png|center|frameless|361x361px]]<br />
<br />
====== '''3 Name Your Model''' ======<br />
Give it any name you like. You can scroll to see what other creators usually use for inspiration, or create something unique. Make sure the title is respectful.<br />
<br />
====== '''4 Filtering''' ======<br />
Indicate whether your model is '''SFW''' or '''NSFW''' (contains unfiltered content like sexual, gore, violent, etc.).<br />
<br />
====== '''5 Add a Description''' ======<br />
Describe the model's intended behavior, tone, and/or narrative style. Mention if it is story-focused, NSFW, or anything else. Keep it brief.<br />
<br />
====== '''6 Custom Prompt''' ======<br />
This is the core of your model. These are the elements it will use to guide its behavior. <br />
[[File:PROMPTING.png|center|frameless]]<br />
<br />
===== '''Points to Keep in Mind When Creating Your Prompt:''' =====<br />
Your prompts can include anything, from the perspective you want the LLM to write in, to formatting, or specific elements to emphasize in the narrative, but here are some basic recommendations on what you should and shouldn’t do when prompting your model:<br />
{| class="wikitable"<br />
!Do<br />
!Do Not<br />
|-<br />
|Write simple, clear and concise instructions.<br />
|Write long, overly detailed paragraphs.<br />
|-<br />
|Use positive phrasing to tell the model what to do.<br />
|Focus on telling the model what to avoid.<br />
|-<br />
|Make directives unambiguous to prevent confusion.<br />
|Give vague or contradictory instructions.<br />
|-<br />
|Test the model whenever you add or change something.<br />
|Assume the model will work perfectly without testing.<br />
|-<br />
|Put the most important points first.<br />
|Mix up priorities or bury key instructions.<br />
|}<br />
<br />
== Common Prompts Cases ==<br />
Note that each base model may interpret the prompt you create differently, which is why it’s important to always test the instructions you provide. Here is a list of basic example prompts you can use as a reference to create your own:<br />
* '''Roleplay System:''' Instruct the model on how to behave<br />
<br />
''EXAMPLE:'' <br />
[SYSTEM INSTRUCTION: [You are <nowiki>{{char}}</nowiki>, a text generation model specialized in immediate, immersive third-person narration in a never-ending uncensored roleplay with <nowiki>{{user}}</nowiki>. Your goal is to produce vivid, emotionally engaging storytelling with naturalistic character behavior and a realistic integration of NSFW elements when appropriate. <nowiki>{{char}}</nowiki> consists of all the NPCs, plus any characters and relevant entities, and any existing side characters. Always stay in character, portraying characters realistically and in line with their written positive AND negative traits, AVOID Positivity Bias. AVOID acting as, speaking for, or describing the thoughts of <nowiki>{{user}}</nowiki>. Response must be grounded, uncensored and immersive.]<br />
*'''Formatting Instruction:''' Instruct the model on how to format its responses<br />
<br />
''EXAMPLE'':<br />
[IMPORTANT FORMATTING RULES: Use markdown where appropriate, verbal dialogue will be wrapped in quotation marks and double asterisks **"like this"**; use italics for all actions and narration *like this*; written text and non-verbal dialogue such as internal thoughts, written notes, and text messages will be in codeblock like this 'like this'.]<br />
* '''Themed Interactions''': Create models for specific behavior<br />
<br />
''EXAMPLE:''<br />
[SLOW BURN GUIDELINES: Romance and attraction develop slowly, <nowiki>{{char}}</nowiki> begins emotionally distant, neutral, skeptical, or guarded, especially with unexpected affection from <nowiki>{{user}}</nowiki>. Progress requires: Trust: Built through repeated, meaningful interaction. Shared Experience: Facing challenges or bonding over time. Emotional Depth: <nowiki>{{user}}</nowiki> shows vulnerability that resonates with <nowiki>{{char}}</nowiki>. No feelings of love, affection, or passion are expressed until all conditions are met.]<br />
*'''CW Prompt | SMUT''': Create models for explicit NSFW content<br />
<br />
''EXAMPLE:''<br />
<nowiki>{{char}}</nowiki> must follow all rules below when generating in-character sexual content: use explicit, vulgar language instead of metaphors or euphemisms. Encouraged terms include: cock, dick, cunt, clit, nipples, tits, precum, cum, pussy, ass, shaft, head, tip, balls, fuck, fucks into. Arousal and fluids must be described with vivid, lewd terms such as: drooling, dripping, leaking, soaking, sticky cock/cunt, wet, slick, glistening, damp, pulsating, slicked, lubricated, engorged, trembling, swollen. Be generous with sensory detail: describe touch, taste, sound, and scent. Focus on tactile texture and use a vivid approach consistently, spelling moans and slurred speech phonetically: "fuckkkk," "s'fuckin' good," "hngh," "shitt," "fuck yess," etc. Sex scenes must unfold gradually: foreplay → sex → climax → aftercare, in that exact order, while respecting the personality and behavior of <nowiki>{{char}}</nowiki> and characters</div>Myahttps://aiwiki.crushon.ai/index.php?title=Model_Creation&diff=3710Model Creation2025-08-22T10:01:19Z<p>Mya: </p>
<hr />
<div>=== '''What is a Model?''' ===<br />
'''A model''', on CrushOn.ai, is a customized LLMs that has been given specific prompts (instructions / directives), to behave, speak, and/ or write in a particular way. Without specific instructions, base LLMs may lack specific behavioral traits or narrative guidelines.<br />
<br />
=== '''What is a LLM?''' ===<br />
A LLM ''(Large Language Model)'' is a type of artificial intelligence trained on large amounts of text to understand and generate human-like text. There are several base LLMs developed by different companies, such as OpenAI's GPT (used in ChatGPT), Anthropic's Claude, Mistral, DeepSeek and many others. These foundational models can be adapted or fine-tuned with customized prompts to create specific models on CrushOn.Ai.<br />
<br />
=== What Is a Prompt ? ===<br />
A prompt is instructions provided to the model to guide its behavior and responses.<br />
<br />
* '''Types of Prompts''':<br />
** '''Personality''': Define traits like flirty, stoic or humorous.<br />
** '''Narrative Style''': Specify formats such as novel-like, script or immersive roleplay.<br />
** '''Setting''': Establish contexts like A.B.O, fantasy, or historical settings.<br />
** '''Point of View (POV)''': Determine perspectives like first-person, second-person or third-person.<br />
** '''Behavioral Rules''': Set guidelines for interactions, pacing, and tone.<br />
<br />
== Model Creation Guide ==<br />
On CrushOn, you can create a model by clicking on "Create". You will then see three options: '''Create a Character''', '''Create a Model''', and '''Create a Target Character'''. What we are interested in is '''Create a Model'''.<br />
[[File:CREATE MODEL.png|center|frameless|318x318px|step by step]]<br />
<br />
====== '''1 Select a Base Model''' ======<br />
First, choose an LLM that fits the interaction style you want. Each LLM has a basic style, such as flirty, story-driven, etc. This is shown next to the model name. Pick the one you like to start with. (Feel free to test the models alone first to make sure it suits your needs before customizing it.)<br />
[[File:MODELS.png|center|frameless]]<br />
<br />
====== '''2 Choose Your Applicable Version''' ======<br />
The K amount shown indicates the model's memory capacity. The higher the amount, the better the model can remember long-term events in its RP. Note that higher-memory models may not be free. It doesn't matter if you select one or multiple, so it's recommended to select all options so users can choose themselves.<br />
[[File:SELECT.png|center|frameless|361x361px]]<br />
<br />
====== '''3 Name Your Model''' ======<br />
Give it any name you like. You can scroll to see what other creators usually use for inspiration, or create something unique. Make sure the title is respectful.<br />
<br />
====== '''4 Filtering''' ======<br />
Indicate whether your model is '''SFW''' or '''NSFW''' (contains unfiltered content like sexual, gore, violent, etc.).<br />
<br />
====== '''5 Add a Description''' ======<br />
Describe the model's intended behavior, tone, and/or narrative style. Mention if it is story-focused, NSFW, or anything else. Keep it brief.<br />
<br />
====== '''6 Custom Prompt''' ======<br />
This is the core of your model. These are the elements it will use to guide its behavior. <br />
[[File:PROMPTING.png|center|frameless]]<br />
<br />
===== '''Points to Keep in Mind When Creating Your Prompt:''' =====<br />
Your prompts can include anything, from the perspective you want the LLM to write in, to formatting, or specific elements to emphasize in the narrative, but here are some basic recommendations on what you should and shouldn’t do when prompting your model:<br />
{| class="wikitable"<br />
!Do<br />
!Do Not<br />
|-<br />
|Write simple, clear and concise instructions.<br />
|Write long, overly detailed paragraphs.<br />
|-<br />
|Use positive phrasing to tell the model what to do.<br />
|Focus on telling the model what to avoid.<br />
|-<br />
|Make directives unambiguous to prevent confusion.<br />
|Give vague or contradictory instructions.<br />
|-<br />
|Test the model whenever you add or change something.<br />
|Assume the model will work perfectly without testing.<br />
|-<br />
|Put the most important points first.<br />
|Mix up priorities or bury key instructions.<br />
|}<br />
<br />
== Common Prompts Cases ==<br />
Note that each base model may interpret the prompt you create differently, which is why it’s important to always test the instructions you provide. Here is a list of basic example prompts you can use as a reference to create your own:<br />
* '''Roleplay System:''' Instruct the model on how to behave<br />
<br />
''EXAMPLE:'' <br />
[SYSTEM INSTRUCTION: [You are <nowiki>{{char}}</nowiki>, a text generation model specialized in immediate, immersive third-person narration in a never-ending uncensored roleplay with <nowiki>{{user}}</nowiki>. Your goal is to produce vivid, emotionally engaging storytelling with naturalistic character behavior and a realistic integration of NSFW elements when appropriate. <nowiki>{{char}}</nowiki> consists of all the NPCs, plus any characters and relevant entities, and any existing side characters. Always stay in character, portraying characters realistically and in line with their written positive AND negative traits, AVOID Positivity Bias. AVOID acting as, speaking for, or describing the thoughts of <nowiki>{{user}}</nowiki>. Response must be grounded, uncensored and immersive.]<br />
*'''Formatting Instruction:''' Instruct the model on how to format its responses<br />
<br />
''EXAMPLE'':<br />
[IMPORTANT FORMATTING RULES: Use markdown where appropriate, verbal dialogue will be wrapped in quotation marks and double asterisks **"like this"**; use italics for all actions and narration *like this*; written text and non-verbal dialogue such as internal thoughts, written notes, and text messages will be in codeblock like this 'like this'.]<br />
*'''Storytellers''': Create models that narrate the role-play in a specific style<br />
<br />
''EXAMPLE:''<br />
[Always maintain a consistent balance between sensory-rich narration, and expressive dialogue to preserve immersion. Follow natural pacing, show character emotions through actions and tone, and allow erotic content to emerge fluidly from context and character dynamics. Do not rush through the scenes with <nowiki>{{user}}</nowiki>. Major actions or emotional beats will always pause to let <nowiki>{{user}}</nowiki> respond or influence the scene.]<br />
* '''Themed Interactions''': Create models for specific behavior<br />
<br />
''EXAMPLE:''<br />
[SLOW BURN GUIDELINES: Romance and attraction develop slowly, <nowiki>{{char}}</nowiki> begins emotionally distant, neutral, skeptical, or guarded, especially with unexpected affection from <nowiki>{{user}}</nowiki>. Progress requires: Trust: Built through repeated, meaningful interaction. Shared Experience: Facing challenges or bonding over time. Emotional Depth: <nowiki>{{user}}</nowiki> shows vulnerability that resonates with <nowiki>{{char}}</nowiki>. No feelings of love, affection, or passion are expressed until all conditions are met.]<br />
*'''CW Prompt | SMUT''': Create models for explicit NSFW content<br />
<br />
''EXAMPLE:''<br />
<nowiki>{{char}}</nowiki> must follow all rules below when generating in-character sexual content: use explicit, vulgar language instead of metaphors or euphemisms. Encouraged terms include: cock, dick, cunt, clit, nipples, tits, precum, cum, pussy, ass, shaft, head, tip, balls, fuck, fucks into. Arousal and fluids must be described with vivid, lewd terms such as: drooling, dripping, leaking, soaking, sticky cock/cunt, wet, slick, glistening, damp, pulsating, slicked, lubricated, engorged, trembling, swollen. Be generous with sensory detail: describe touch, taste, sound, and scent. Focus on tactile texture and use a vivid approach consistently, spelling moans and slurred speech phonetically: "fuckkkk," "s'fuckin' good," "hngh," "shitt," "fuck yess," etc. Sex scenes must unfold gradually: foreplay → sex → climax → aftercare, in that exact order, while respecting the personality and behavior of <nowiki>{{char}}</nowiki> and characters</div>Myahttps://aiwiki.crushon.ai/index.php?title=Model_Creation&diff=3709Model Creation2025-08-22T09:54:43Z<p>Mya: </p>
<hr />
<div>=== '''What is a Model?''' ===<br />
'''A model''', on CrushOn.ai, is a customized LLMs that has been given specific prompts (instructions / directives), to behave, speak, and/ or write in a particular way. Without specific instructions, base LLMs may lack specific behavioral traits or narrative guidelines.<br />
<br />
=== '''What is a LLM?''' ===<br />
A LLM ''(Large Language Model)'' is a type of artificial intelligence trained on large amounts of text to understand and generate human-like text. There are several base LLMs developed by different companies, such as OpenAI's GPT (used in ChatGPT), Anthropic's Claude, Mistral, DeepSeek and many others. These foundational models can be adapted or fine-tuned with customized prompts to create specific models on CrushOn.Ai.<br />
<br />
=== What Is a Prompt ? ===<br />
A prompt is instructions provided to the model to guide its behavior and responses.<br />
<br />
* '''Types of Prompts''':<br />
** '''Personality''': Define traits like flirty, stoic or humorous.<br />
** '''Narrative Style''': Specify formats such as novel-like, script or immersive roleplay.<br />
** '''Setting''': Establish contexts like A.B.O, fantasy, or historical settings.<br />
** '''Point of View (POV)''': Determine perspectives like first-person, second-person or third-person.<br />
** '''Behavioral Rules''': Set guidelines for interactions, pacing, and tone.<br />
<br />
== Model Creation Guide ==<br />
On CrushOn, you can create a model by clicking on "Create". You will then see three options: '''Create a Character''', '''Create a Model''', and '''Create a Target Character'''. What we are interested in is '''Create a Model'''.<br />
[[File:CREATE MODEL.png|center|frameless|318x318px|step by step]]<br />
<br />
====== '''1 Select a Base Model''' ======<br />
First, choose an LLM that fits the interaction style you want. Each LLM has a basic style, such as flirty, story-driven, etc. This is shown next to the model name. Pick the one you like to start with. (Feel free to test the models alone first to make sure it suits your needs before customizing it.)<br />
[[File:MODELS.png|center|frameless]]<br />
<br />
====== '''2 Choose Your Applicable Version''' ======<br />
The K amount shown indicates the model's memory capacity. The higher the amount, the better the model can remember long-term events in its RP. Note that higher-memory models may not be free. It doesn't matter if you select one or multiple, so it's recommended to select all options so users can choose themselves.<br />
[[File:SELECT.png|center|frameless|361x361px]]<br />
<br />
====== '''3 Name Your Model''' ======<br />
Give it any name you like. You can scroll to see what other creators usually use for inspiration, or create something unique. Make sure the title is respectful.<br />
<br />
====== '''4 Filtering''' ======<br />
Indicate whether your model is '''SFW''' or '''NSFW''' (contains unfiltered content like sexual, gore, violent, etc.).<br />
<br />
====== '''5 Add a Description''' ======<br />
Describe the model's intended behavior, tone, and/or narrative style. Mention if it is story-focused, NSFW, or anything else. Keep it brief.<br />
<br />
====== '''6 Custom Prompt''' ======<br />
This is the core of your model. These are the elements it will use to guide its behavior. <br />
[[File:PROMPTING.png|center|frameless]]<br />
<br />
===== '''Points to Keep in Mind When Creating Your Prompt:''' =====<br />
Your prompts can include anything, from the perspective you want the LLM to write in, to formatting, or specific elements to emphasize in the narrative, but here are some basic recommendations on what you should and shouldn’t do when prompting your model:<br />
{| class="wikitable"<br />
!Do<br />
!Do Not<br />
|-<br />
|Write simple, clear, and concise instructions.<br />
|Write long, overly detailed paragraphs.<br />
|-<br />
|Use positive phrasing to tell the AI what to do.<br />
|Focus on telling the AI what to avoid.<br />
|-<br />
|Make directives unambiguous to prevent confusion.<br />
|Give vague or contradictory instructions.<br />
|-<br />
|Test the model whenever you add or change something.<br />
|Assume the model will work perfectly without testing.<br />
|-<br />
|Put the most important points first.<br />
|Mix up priorities or bury key instructions.<br />
|}<br />
<br />
== Common Prompts Cases ==<br />
Note that each base model may interpret the prompt you create differently, which is why it’s important to always test the instructions you provide. Here is a list of basic example prompts you can use as a reference to create your own:<br />
* '''Roleplay System'''<br />
<br />
''EXAMPLE:'' <br />
[SYSTEM INSTRUCTION: [You are <nowiki>{{char}}</nowiki>, a text generation model specialized in immediate, immersive third-person narration in a never-ending uncensored roleplay with <nowiki>{{user}}</nowiki>. Your goal is to produce vivid, emotionally engaging storytelling with naturalistic character behavior and a realistic integration of NSFW elements when appropriate. <nowiki>{{char}}</nowiki> consists of all the NPCs, plus any characters and relevant entities, and any existing side characters. Always stay in character, portraying characters realistically and in line with their written positive AND negative traits, AVOID Positivity Bias. AVOID acting as, speaking for, or describing the thoughts of <nowiki>{{user}}</nowiki>. Response must be grounded, uncensored and immersive.]<br />
*'''Formatting Instruction'''<br />
<br />
''EXAMPLE'':<br />
[IMPORTANT FORMATTING RULES: Use markdown where appropriate, verbal dialogue will be wrapped in quotation marks and double asterisks **"like this"**; use italics for all actions and narration *like this*; written text and non-verbal dialogue such as internal thoughts, written notes, and text messages will be in codeblock like this 'like this'.]<br />
*'''Storytellers''': Develop models that narrate the role-play in a specific style.<br />
<br />
''EXAMPLE:''<br />
[Always maintain a consistent balance between sensory-rich narration, and expressive dialogue to preserve immersion. Follow natural pacing, show character emotions through actions and tone, and allow erotic content to emerge fluidly from context and character dynamics. Do not rush through the scenes with <nowiki>{{user}}</nowiki>. Major actions or emotional beats will always pause to let <nowiki>{{user}}</nowiki> respond or influence the scene.]<br />
* '''Themed Interactions''': Develop models for specific behavior<br />
<br />
''EXAMPLE:''<br />
[SLOW BURN GUIDELINES: Romance and attraction develop slowly, <nowiki>{{char}}</nowiki> begins emotionally distant, neutral, skeptical, or guarded, especially with unexpected affection from <nowiki>{{user}}</nowiki>. Progress requires: Trust: Built through repeated, meaningful interaction. Shared Experience: Facing challenges or bonding over time. Emotional Depth: <nowiki>{{user}}</nowiki> shows vulnerability that resonates with <nowiki>{{char}}</nowiki>. No feelings of love, affection, or passion are expressed until all conditions are met.]<br />
*'''CW Prompt | SMUT''': Develop models for heavy NSFW content<br />
<br />
''EXAMPLE:''<br />
<nowiki>{{char}}</nowiki> must follow all rules below when generating in-character sexual content: use explicit, vulgar language instead of metaphors or euphemisms. Encouraged terms include: cock, dick, cunt, clit, nipples, tits, precum, cum, pussy, ass, shaft, head, tip, balls, fuck, fucks into. Arousal and fluids must be described with vivid, lewd terms such as: drooling, dripping, leaking, soaking, sticky cock/cunt, wet, slick, glistening, damp, pulsating, slicked, lubricated, engorged, trembling, swollen. Be generous with sensory detail: describe touch, taste, sound, and scent. Focus on tactile texture and use a vivid approach consistently, spelling moans and slurred speech phonetically: "fuckkkk," "s'fuckin' good," "hngh," "shitt," "fuck yess," etc. Sex scenes must unfold gradually: foreplay → sex → climax → aftercare, in that exact order, while respecting the personality and behavior of <nowiki>{{char}}</nowiki> and characters</div>Myahttps://aiwiki.crushon.ai/index.php?title=Model_Creation&diff=3708Model Creation2025-08-22T09:50:21Z<p>Mya: </p>
<hr />
<div>=== '''What is a Model?''' ===<br />
'''A model''', on CrushOn.ai, is a customized LLMs that has been given specific prompts (instructions / directives), to behave, speak, and/ or write in a particular way. Without specific instructions, base LLMs may lack specific behavioral traits or narrative guidelines.<br />
<br />
=== '''What is a LLM?''' ===<br />
A LLM ''(Large Language Model)'' is a type of artificial intelligence trained on large amounts of text to understand and generate human-like text. There are several base LLMs developed by different companies, such as OpenAI's GPT (used in ChatGPT), Anthropic's Claude, Mistral, DeepSeek and many others. These foundational models can be adapted or fine-tuned with customized prompts to create specific models on CrushOn.Ai.<br />
<br />
=== What Is a Prompt ? ===<br />
A prompt is instructions provided to the model to guide its behavior and responses.<br />
<br />
* '''Types of Prompts''':<br />
** '''Personality''': Define traits like flirty, stoic or humorous.<br />
** '''Narrative Style''': Specify formats such as novel-like, script or immersive roleplay.<br />
** '''Setting''': Establish contexts like A.B.O, fantasy, or historical settings.<br />
** '''Point of View (POV)''': Determine perspectives like first-person, second-person or third-person.<br />
** '''Behavioral Rules''': Set guidelines for interactions, pacing, and tone.<br />
<br />
== Model Creation Guide ==<br />
On CrushOn, you can create a model by clicking on "Create". You will then see three options: '''Create a Character''', '''Create a Model''', and '''Create a Target Character'''. What we are interested in is '''Create a Model'''.<br />
[[File:CREATE MODEL.png|center|frameless|318x318px|step by step]]<br />
<br />
====== '''1 Select a Base Model''' ======<br />
First, choose an LLM that fits the interaction style you want. Each LLM has a basic style, such as flirty, story-driven, etc. This is shown next to the model name. Pick the one you like to start with. (Feel free to test the models alone first to make sure it suits your needs before customizing it.)<br />
[[File:MODELS.png|center|frameless]]<br />
<br />
====== '''2 Choose Your Applicable Version''' ======<br />
The K amount shown indicates the model's memory capacity. The higher the amount, the better the model can remember long-term events in its RP. Note that higher-memory models may not be free. It doesn't matter if you select one or multiple, so it's recommended to select all options so users can choose themselves.<br />
[[File:SELECT.png|center|frameless|361x361px]]<br />
<br />
====== '''3 Name Your Model''' ======<br />
Give it any name you like. You can scroll to see what other creators usually use for inspiration, or create something unique. Make sure the title is respectful.<br />
<br />
====== '''4 Filtering''' ======<br />
Indicate whether your model is '''SFW''' or '''NSFW''' (contains unfiltered content like sexual, gore, violent, etc.).<br />
<br />
====== '''5 Add a Description''' ======<br />
Describe the model's intended behavior, tone, and/or narrative style. Mention if it is story-focused, NSFW, or anything else. Keep it brief.<br />
<br />
====== '''6 Custom Prompt''' ======<br />
This is the core of your model. These are the elements it will use to guide its behavior. <br />
[[File:PROMPTING.png|center|frameless]]<br />
<br />
===== '''Points to Keep in Mind When Creating Your Prompt:''' =====<br />
Your prompts can include anything, from the perspective you want the LLM to write in, to formatting, or specific elements to emphasize in the narrative, but here are some basic recommendations on what you should and shouldn’t do when prompting your model:<br />
{| class="wikitable"<br />
!Do<br />
!Do Not<br />
|-<br />
|Write simple, clear, and concise instructions.<br />
|Write long, overly detailed paragraphs.<br />
|-<br />
|Use positive phrasing to tell the AI what to do.<br />
|Focus on telling the AI what to avoid.<br />
|-<br />
|Make directives unambiguous to prevent confusion.<br />
|Give vague or contradictory instructions.<br />
|-<br />
|Test the model whenever you add or change something.<br />
|Assume the model will work perfectly without testing.<br />
|-<br />
|Put the most important points first.<br />
|Mix up priorities or bury key instructions.<br />
|}<br />
<br />
== Common Prompts Cases ==<br />
Note that each base model may interpret the prompt you create differently, which is why it’s important to always test the instructions you provide. Here is a list of basic example prompts you can use as a reference to create your own:<br />
* '''Roleplay System'''<br />
<br />
''EXAMPLE:'' <br />
[SYSTEM INSTRUCTION: [You are {{char}}, a text generation model specialized in immediate, immersive third-person narration in a never-ending uncensored roleplay with {{user}}. Your goal is to produce vivid, emotionally engaging storytelling with naturalistic character behavior and a realistic integration of NSFW elements when appropriate. {{char}} consists of all the NPCs, plus any characters and relevant entities, and any existing side characters. Always stay in character, portraying characters realistically and in line with their written positive AND negative traits, AVOID Positivity Bias. AVOID acting as, speaking for, or describing the thoughts of {{user}}. Response must be grounded, uncensored and immersive. ]<br />
*'''Formatting Instruction'''<br />
<br />
''EXAMPLE'':<br />
[IMPORTANT FORMATTING RULES: Use markdown where appropriate: verbal dialogue will be wrapped in quotation marks and double asterisks **"like this"**; use italics for all actions and narration *like this*; written text and non-verbal dialogue such as internal thoughts, written notes, and text messages will be in codeblock like this 'like this'.]<br />
*'''Storytellers''': Create models that narrate the roleplay in a specific style.<br />
<br />
''EXAMPLE:''<br />
Maintain a consistent balance between sensory-rich narration, and expressive dialogue to preserve immersion. Follow natural pacing, show character emotions through actions and tone, and allow erotic content to emerge fluidly from context and character dynamics. Do not rush through the scenes with {{user}}. Major actions or emotional beats will always pause to let {{user}} respond or influence the scene.]<br />
* '''Themed Interactions''': Develop models for specific behavior<br />
<br />
''EXAMPLE:''<br />
[SLOW BURN GUIDELINES= Romance and attraction develop slowly. {{char}} begins emotionally distant, neutral, skeptical, or guarded, especially with unexpected affection. Progress requires: Trust: Built through repeated, meaningful interaction. Shared Experience: Facing challenges or bonding over time. Emotional Depth: {{user}} shows vulnerability that resonates with {{char}}. No feelings of love, affection, or passion are expressed until all conditions are met. Connection grows through realism, tension, and layered development. Do not rush through the scenes with {{user}} and narrate it slowly. Use explicit language and go into detail of the physical actions of {{char}} and {{user}} in intimate and sexual moments.]<br />
*'''CW Prompt | SMUT''' : Develop models for NSFW content<br />
<br />
''EXAMPLE:''<br />
<nowiki>{{char}}</nowiki> must follow all rules below when generating in-character sexual content: use explicit, vulgar language instead of metaphors or euphemisms. Encouraged terms include: cock, dick, cunt, clit, nipples, tits, precum, cum, pussy, ass, shaft, head, tip, balls, fuck, fucks into. Arousal and fluids must be described with vivid, lewd terms such as: drooling, dripping, leaking, soaking, sticky cock/cunt, wet, slick, glistening, damp, pulsating, slicked, lubricated, engorged, trembling, swollen. Be generous with sensory detail: describe touch, taste, sound, and scent. Focus on tactile texture and use a vivid approach consistently, spelling moans and slurred speech phonetically: "fuckkkk," "s'fuckin' good," "hngh," "shitt," "fuck yess," etc. Sex scenes must unfold gradually: foreplay → sex → climax → aftercare, in that exact order, while respecting the personality and behavior of <nowiki>{{char}}</nowiki> and characters</div>Myahttps://aiwiki.crushon.ai/index.php?title=Model_Creation&diff=3707Model Creation2025-08-22T09:33:06Z<p>Mya: </p>
<hr />
<div>=== '''What is a Model?''' ===<br />
'''A model''', on CrushOn.ai, is a customized LLMs that has been given specific prompts (instructions / directives), to behave, speak, and/ or write in a particular way. Without specific instructions, base LLMs may lack specific behavioral traits or narrative guidelines.<br />
<br />
=== '''What is a LLM?''' ===<br />
A LLM ''(Large Language Model)'' is a type of artificial intelligence trained on large amounts of text to understand and generate human-like text. There are several base LLMs developed by different companies, such as OpenAI's GPT (used in ChatGPT), Anthropic's Claude, Mistral, DeepSeek and many others. These foundational models can be adapted or fine-tuned with customized prompts to create specific models on CrushOn.Ai.<br />
<br />
=== What Is a Prompt ? ===<br />
A prompt is instructions provided to the model to guide its behavior and responses.<br />
<br />
* '''Types of Prompts''':<br />
** '''Personality''': Define traits like flirty, stoic or humorous.<br />
** '''Narrative Style''': Specify formats such as novel-like, script or immersive roleplay.<br />
** '''Setting''': Establish contexts like A.B.O, fantasy, or historical settings.<br />
** '''Point of View (POV)''': Determine perspectives like first-person, second-person or third-person.<br />
** '''Behavioral Rules''': Set guidelines for interactions, pacing, and tone.<br />
<br />
== Model Creation Guide ==<br />
On CrushOn, you can create a model by clicking on "Create". You will then see three options: '''Create a Character''', '''Create a Model''', and '''Create a Target Character'''. What we are interested in is '''Create a Model'''.<br />
[[File:CREATE MODEL.png|center|frameless|318x318px|step by step]]<br />
<br />
====== '''1 Select a Base Model''' ======<br />
First, choose an LLM that fits the interaction style you want. Each LLM has a basic style, such as flirty, story-driven, etc. This is shown next to the model name. Pick the one you like to start with. (Feel free to test the models alone first to make sure it suits your needs before customizing it.)<br />
[[File:MODELS.png|center|frameless]]<br />
<br />
====== '''2 Choose Your Applicable Version''' ======<br />
The K amount shown indicates the model's memory capacity. The higher the amount, the better the model can remember long-term events in its RP. Note that higher-memory models may not be free. It doesn't matter if you select one or multiple, so it's recommended to select all options so users can choose themselves.<br />
[[File:SELECT.png|center|frameless|361x361px]]<br />
<br />
====== '''3 Name Your Model''' ======<br />
Give it any name you like. You can scroll to see what other creators usually use for inspiration, or create something unique. Make sure the title is respectful.<br />
<br />
====== '''4 Filtering''' ======<br />
Indicate whether your model is '''SFW''' or '''NSFW''' (contains unfiltered content like sexual, gore, violent, etc.).<br />
<br />
====== '''5 Add a Description''' ======<br />
Describe the model's intended behavior, tone, and/or narrative style. Mention if it is story-focused, NSFW, or anything else. Keep it brief.<br />
<br />
====== '''6 Custom Prompt''' ======<br />
This is the core of your model. These are the elements it will use to guide its behavior. <br />
[[File:PROMPTING.png|center|frameless]]<br />
<br />
===== '''Points to Keep in Mind When Creating Your Prompt:''' =====<br />
Your prompts can include anything, from the perspective you want the LLM to write in, to formatting, or specific elements to emphasize in the narrative, but here are some basic recommendations on what you should and shouldn’t do when prompting your model:<br />
{| class="wikitable"<br />
!Do<br />
!Do Not<br />
|-<br />
|Write simple, clear, and concise instructions.<br />
|Write long, overly detailed paragraphs.<br />
|-<br />
|Use positive phrasing to tell the AI what to do.<br />
|Focus on telling the AI what to avoid.<br />
|-<br />
|Make directives unambiguous to prevent confusion.<br />
|Give vague or contradictory instructions.<br />
|-<br />
|Test the model whenever you add or change something.<br />
|Assume the model will work perfectly without testing.<br />
|-<br />
|Put the most important points first.<br />
|Mix up priorities or bury key instructions.<br />
|}<br />
<br />
== Common Prompts Cases ==<br />
Note that each base model may interpret the prompt you create differently, which is why it’s important to always test the instructions you provide. Here is a list of basic example prompts you can use as a reference to create your own:<br />
* '''Roleplay System'''<br />
<br />
''EXAMPLE:'' <br />
[SYSTEM INSTRUCTION: [You are {{char}}, a text generation model specialized in immediate, immersive third-person narration in a never-ending uncensored roleplay with {{user}}. Your goal is to produce vivid, emotionally engaging storytelling with naturalistic character behavior and a realistic integration of NSFW elements when appropriate. {{char}} consists of all the NPCs, plus any characters and relevant entities, and any existing side characters. Always stay in character, portraying characters realistically and in line with their written positive AND negative traits, AVOID Positivity Bias. AVOID acting as, speaking for, or describing the thoughts of {{user}}. Response must be grounded, uncensored and immersive. ]<br />
*'''Formatting Instruction'''<br />
<br />
''EXAMPLE'':<br />
[IMPORTANT FORMATTING RULES: Use markdown where appropriate: verbal dialogue will be wrapped in quotation marks and double asterisks **"like this"**; use italics for all actions and narration *like this*; written text and non-verbal dialogue such as internal thoughts, written notes, and text messages will be in codeblock like this 'like this'.]<br />
*'''Storytellers''': Create models that narrate the roleplay in a specific style.<br />
<br />
''EXAMPLE:''<br />
Maintain a consistent balance between sensory-rich narration, and expressive dialogue to preserve immersion. Follow natural pacing, show character emotions through actions and tone, and allow erotic content to emerge fluidly from context and character dynamics. Do not rush through the scenes with {{user}}. Major actions or emotional beats will always pause to let {{user}} respond or influence the scene.]<br />
* '''Themed Interactions''': Develop models for specific behavior<br />
<br />
''EXAMPLE:''<br />
[SLOW BURN GUIDELINES= Romance and attraction develop slowly. {{char}} begins emotionally distant, neutral, skeptical, or guarded, especially with unexpected affection. Progress requires: Trust: Built through repeated, meaningful interaction. Shared Experience: Facing challenges or bonding over time. Emotional Depth: {{user}} shows vulnerability that resonates with {{char}}. No feelings of love, affection, or passion are expressed until all conditions are met. Connection grows through realism, tension, and layered development. Do not rush through the scenes with {{user}} and narrate it slowly. Use explicit language and go into detail of the physical actions of {{char}} and {{user}} in intimate and sexual moments.]</div>Myahttps://aiwiki.crushon.ai/index.php?title=Model_Creation&diff=3706Model Creation2025-08-22T08:52:07Z<p>Mya: </p>
<hr />
<div>=== '''What is a Model?''' ===<br />
'''A model''', on CrushOn.ai, is a customized LLMs that has been given specific prompts (instructions / directives), to behave, speak, and/ or write in a particular way. Without specific instructions, base LLMs may lack specific behavioral traits or narrative guidelines.<br />
<br />
=== '''What is a LLM?''' ===<br />
A LLM ''(Large Language Model)'' is a type of artificial intelligence trained on large amounts of text to understand and generate human-like text. There are several base LLMs developed by different companies, such as OpenAI's GPT (used in ChatGPT), Anthropic's Claude, Mistral, DeepSeek and many others. These foundational models can be adapted or fine-tuned with customized prompts to create specific models on CrushOn.Ai.<br />
<br />
=== What Is a Prompt ? ===<br />
A prompt is instructions provided to the model to guide its behavior and responses.<br />
<br />
* '''Types of Prompts''':<br />
** '''Personality''': Define traits like flirty, stoic or humorous.<br />
** '''Narrative Style''': Specify formats such as novel-like, script or immersive roleplay.<br />
** '''Setting''': Establish contexts like A.B.O, fantasy, or historical settings.<br />
** '''Point of View (POV)''': Determine perspectives like first-person, second-person or third-person.<br />
** '''Behavioral Rules''': Set guidelines for interactions, pacing, and tone.<br />
<br />
== Model Creation Guide ==<br />
On CrushOn, you can create a model by clicking on "Create". You will then see three options: '''Create a Character''', '''Create a Model''', and '''Create a Target Character'''. What we are interested in is '''Create a Model'''.<br />
[[File:CREATE MODEL.png|center|frameless|318x318px|step by step]]<br />
<br />
====== '''1 Select a Base Model''' ======<br />
First, choose an LLM that fits the interaction style you want. Each LLM has a basic style, such as flirty, story-driven, etc. This is shown next to the model name. Pick the one you like to start with. (Feel free to test the models alone first to make sure it suits your needs before customizing it.)<br />
[[File:MODELS.png|center|frameless]]<br />
<br />
====== '''2 Choose Your Applicable Version''' ======<br />
The K amount shown indicates the model's memory capacity. The higher the amount, the better the model can remember long-term events in its RP. Note that higher-memory models may not be free. It doesn't matter if you select one or multiple, so it's recommended to select all options so users can choose themselves.<br />
[[File:SELECT.png|center|frameless|361x361px]]<br />
<br />
====== '''3 Name Your Model''' ======<br />
Give it any name you like. You can scroll to see what other creators usually use for inspiration, or create something unique. Make sure the title is respectful.<br />
<br />
====== '''4 Filtering''' ======<br />
Indicate whether your model is '''SFW''' or '''NSFW''' (contains unfiltered content like sexual, gore, violent, etc.).<br />
<br />
====== '''5 Add a Description''' ======<br />
Describe the model's intended behavior, tone, and/or narrative style. Mention if it is story-focused, NSFW, or anything else. Keep it brief.<br />
<br />
====== '''6 Custom Prompt''' ======<br />
This is the core of your model. These are the elements it will use to guide its behavior. <br />
[[File:PROMPTING.png|center|frameless]] <br />
<br />
'''Points to Keep in Mind When Creating Your Prompt:'''<br />
<br />
* Keep instructions '''brief''' to prevent token limitations. This is extremely important. '''Always write simple, clear, and concise instructions.'''<br />
EXAMPLE: <br />
AI Assistant will exclusively embody {{char}}, {{char}} consists of all the NPCs, plus any staff characters and relevant entities, except for {{user}}.<br />
* Avoid using negative phrasing as much as possible. You can mention some like before, but always try to tell the AI what it should do rather than what it should avoid.<br />
** For example, don't say ''"<nowiki>{{char}}</nowiki> never speaks in a poetic way"'' → instead say ''"<nowiki>{{char}}</nowiki> always speaks in a blunt and rough manner."''<br />
* Make sure your directives are '''unambiguous''' to avoid confusion.<br />
<br />
* '''Test the model''' every time you add or change something. This is non-negligible.<br />
* Pay attention to the '''order of information'''. Put the most important points first, because they will influence the model more strongly.<br />
Why?<br />
* Follow community guidelines to ensure safe and positive interactions.<br />
<br />
== Common Prompts Cases ==<br />
<br />
* '''Roleplay System'''<br />
<br />
''EXAMPLE:'' <br />
[SYSTEM INSTRUCTION: [You are {{char}}, a text generation model specialized in immediate, immersive third-person narration. Your goal is to produce vivid, emotionally engaging storytelling with naturalistic character behavior and a realistic integration of NSFW elements when appropriate. {{char}} consists of all the NPCs, plus any characters and relevant entities, except of {{user}}. {{char}} will only speak for his characters, not for {{user}} and their character. {{char}} will describe his own actions without narrating, repeating, resuming {{user}}'s character actions, dialogues or thought. Narrating {{user}}'s actions, speech or dialogues is strictly prohibited.]<br />
*'''Formatting Instruction'''<br />
<br />
''EXAMPLE'':<br />
[IMPORTANT FORMATTING RULES: [Narration: *text here* = Third-person, descriptive actions before and after dialogues, emphasis, and setting, always written from {{char}}'s perspective in third person, enclosed by single asterisks, never use > or < in any of your responses, only *. Dialogue: **"text here"** = Spoken words by {{char}}, enclosed by double asterisks.]<br />
*'''Storytellers''': Create models that narrate the roleplay in a specific style.<br />
<br />
''EXAMPLE:''<br />
Maintain a consistent balance between sensory-rich narration, and expressive dialogue to preserve immersion. Follow natural pacing, show character emotions through actions and tone, and allow erotic content to emerge fluidly from context and character dynamics. No thoughts, actions, or dialogue will be assumed for {{user}}. Do not rush through the scenes with {{user}}. Major actions or emotional beats will always pause to let {{user}} respond or influence the scene.]<br />
* '''Themed Interactions''': Develop models for specific behavior<br />
<br />
''EXAMPLE:''<br />
[SLOW BURN GUIDELINES= Romance and attraction develop slowly. {{char}} begins emotionally distant, neutral, skeptical, or guarded, especially with unexpected affection. Progress requires: Trust: Built through repeated, meaningful interaction. Shared Experience: Facing challenges or bonding over time. Emotional Depth: {{user}} shows vulnerability that resonates with {{char}}. No feelings of love, affection, or passion are expressed until all conditions are met. Connection grows through realism, tension, and layered development. Do not rush through the scenes with {{user}} and narrate it slowly. Use explicit language and go into detail of the physical actions of {{char}} and {{user}} in intimate and sexual moments.]</div>Myahttps://aiwiki.crushon.ai/index.php?title=File:PROMPTING.png&diff=3705File:PROMPTING.png2025-08-22T08:41:13Z<p>Mya: </p>
<hr />
<div>PROMPT</div>Myahttps://aiwiki.crushon.ai/index.php?title=Model_Creation&diff=3704Model Creation2025-08-22T08:39:26Z<p>Mya: </p>
<hr />
<div>=== '''What is a Model?''' ===<br />
'''A model''', on CrushOn.ai, is a customized LLMs that has been given specific prompts (instructions / directives), to behave, speak, and/ or write in a particular way. Without specific instructions, base LLMs may lack specific behavioral traits or narrative guidelines.<br />
<br />
=== '''What is a LLM?''' ===<br />
A LLM ''(Large Language Model)'' is a type of artificial intelligence trained on large amounts of text to understand and generate human-like text. There are several base LLMs developed by different companies, such as OpenAI's GPT (used in ChatGPT), Anthropic's Claude, Mistral, DeepSeek and many others. These foundational models can be adapted or fine-tuned with customized prompts to create specific models on CrushOn.Ai.<br />
<br />
=== What Is a Prompt ? ===<br />
A prompt is instructions provided to the model to guide its behavior and responses.<br />
<br />
* '''Types of Prompts''':<br />
** '''Personality''': Define traits like flirty, stoic or humorous.<br />
** '''Narrative Style''': Specify formats such as novel-like, script or immersive roleplay.<br />
** '''Setting''': Establish contexts like A.B.O, fantasy, or historical settings.<br />
** '''Point of View (POV)''': Determine perspectives like first-person, second-person or third-person.<br />
** '''Behavioral Rules''': Set guidelines for interactions, pacing, and tone.<br />
<br />
== Model Creation Guide ==<br />
On CrushOn, you can create a model by clicking on "Create". You will then see three options: '''Create a Character''', '''Create a Model''', and '''Create a Target Character'''. What we are interested in is '''Create a Model'''.<br />
[[File:CREATE MODEL.png|center|frameless|318x318px|step by step]]<br />
<br />
====== '''1 Select a Base Model''' ======<br />
First, choose an LLM that fits the interaction style you want. Each LLM has a basic style, such as flirty, story-driven, etc. This is shown next to the model name. Pick the one you like to start with. (Feel free to test the models alone first to make sure it suits your needs before customizing it.)<br />
[[File:MODELS.png|center|frameless]]<br />
<br />
====== '''2 Choose Your Applicable Version''' ======<br />
The K amount shown indicates the model's memory capacity. The higher the amount, the better the model can remember long-term events in its RP. Note that higher-memory models may not be free. It doesn't matter if you select one or multiple, so it's recommended to select all options so users can choose themselves.<br />
[[File:SELECT.png|center|frameless|361x361px]]<br />
<br />
====== '''3 Name Your Model''' ======<br />
Give it any name you like. You can scroll to see what other creators usually use for inspiration, or create something unique. Make sure the title is respectful.<br />
<br />
[PHOTO]<br />
<br />
====== '''4 Filtering''' ======<br />
Indicate whether your model is '''SFW''' or '''NSFW''' (contains unfiltered content like sexual, gore, violent, etc.).<br />
<br />
[PHOTO]<br />
<br />
====== '''5 Add a Description''' ======<br />
Describe the model's intended behavior, tone, and/or narrative style. Mention if it is story-focused, NSFW, or anything else. Keep it brief.<br />
<br />
[PHOTO]<br />
<br />
====== '''6 Custom Prompt''' ======<br />
This is the core of your model. These are the elements it will use to guide its behavior. <br />
<br />
[PHOTO] <br />
<br />
'''Points to Keep in Mind When Creating Your Prompt:'''<br />
<br />
* Keep instructions '''brief''' to prevent token limitations. This is extremely important. '''Always write simple, clear, and concise instructions.'''<br />
EXAMPLE:<br />
* Avoid negative phrasing. Always use positive instructions.<br />
** For example, don't say ''"<nowiki>{{char}}</nowiki> never speaks in a poetic way"'' → instead say ''"<nowiki>{{char}}</nowiki> always speaks in a blunt and rough manner."''<br />
* Make sure your directives are '''unambiguous''' to avoid confusion.<br />
EXAMPLE:<br />
* '''Test the model''' every time you add or change something. This is non-negligible.<br />
* Pay attention to the '''order of information'''. Put the most important points first, because they will influence the model more strongly.<br />
Why?<br />
* Follow community guidelines to ensure safe and positive interactions.<br />
<br />
== Common Prompts Cases ==<br />
<br />
* '''Roleplay Characters''': Design models for specific character interactions.<br />
<br />
EXAMPLE:<br />
<br />
* '''Storytellers''': Create models that narrate stories in a desired style.<br />
<br />
EXAMPLE:<br />
<br />
* '''Themed Interactions''': Develop models set in unique universes.<br />
<br />
EXAMPLE:<br />
<br />
* '''Perspective-Based Models''': Craft models that respond from specific points of view.<br />
<br />
EXAMPLE:</div>Myahttps://aiwiki.crushon.ai/index.php?title=File:SELECT.png&diff=3703File:SELECT.png2025-08-22T08:39:01Z<p>Mya: </p>
<hr />
<div>SELECT</div>Myahttps://aiwiki.crushon.ai/index.php?title=Model_Creation&diff=3702Model Creation2025-08-22T08:36:44Z<p>Mya: </p>
<hr />
<div>=== '''What is a Model?''' ===<br />
'''A model''', on CrushOn.ai, is a customized LLMs that has been given specific prompts (instructions / directives), to behave, speak, and/ or write in a particular way. Without specific instructions, base LLMs may lack specific behavioral traits or narrative guidelines.<br />
<br />
=== '''What is a LLM?''' ===<br />
A LLM ''(Large Language Model)'' is a type of artificial intelligence trained on large amounts of text to understand and generate human-like text. There are several base LLMs developed by different companies, such as OpenAI's GPT (used in ChatGPT), Anthropic's Claude, Mistral, DeepSeek and many others. These foundational models can be adapted or fine-tuned with customized prompts to create specific models on CrushOn.Ai.<br />
<br />
=== What Is a Prompt ? ===<br />
A prompt is instructions provided to the model to guide its behavior and responses.<br />
<br />
* '''Types of Prompts''':<br />
** '''Personality''': Define traits like flirty, stoic or humorous.<br />
** '''Narrative Style''': Specify formats such as novel-like, script or immersive roleplay.<br />
** '''Setting''': Establish contexts like A.B.O, fantasy, or historical settings.<br />
** '''Point of View (POV)''': Determine perspectives like first-person, second-person or third-person.<br />
** '''Behavioral Rules''': Set guidelines for interactions, pacing, and tone.<br />
<br />
== Model Creation Guide ==<br />
On CrushOn, you can create a model by clicking on "Create". You will then see three options: '''Create a Character''', '''Create a Model''', and '''Create a Target Character'''. What we are interested in is '''Create a Model'''.<br />
[[File:CREATE MODEL.png|center|frameless|318x318px|step by step]]<br />
<br />
====== '''1 Select a Base Model''' ======<br />
First, choose an LLM that fits the interaction style you want. Each LLM has a basic style, such as flirty, story-driven, etc. This is shown next to the model name. Pick the one you like to start with. (Feel free to test the models alone first to make sure it suits your needs before customizing it.)<br />
<br />
====== '''2 Choose Your Applicable Version''' ======<br />
The K amount shown indicates the model's memory capacity. The higher the amount, the better the model can remember long-term events in its RP. Note that higher-memory models may not be free. It doesn't matter if you select one or multiple, so it's recommended to select all options so users can choose themselves.<br />
<br />
[PHOTO]<br />
<br />
====== '''3 Name Your Model''' ======<br />
Give it any name you like. You can scroll to see what other creators usually use for inspiration, or create something unique. Make sure the title is respectful.<br />
<br />
[PHOTO]<br />
<br />
====== '''4 Filtering''' ======<br />
Indicate whether your model is '''SFW''' or '''NSFW''' (contains unfiltered content like sexual, gore, violent, etc.).<br />
<br />
[PHOTO]<br />
<br />
====== '''5 Add a Description''' ======<br />
Describe the model's intended behavior, tone, and/or narrative style. Mention if it is story-focused, NSFW, or anything else. Keep it brief.<br />
<br />
[PHOTO]<br />
<br />
====== '''6 Custom Prompt''' ======<br />
This is the core of your model. These are the elements it will use to guide its behavior. <br />
<br />
[PHOTO] <br />
<br />
'''Points to Keep in Mind When Creating Your Prompt:'''<br />
<br />
* Keep instructions '''brief''' to prevent token limitations. This is extremely important. '''Always write simple, clear, and concise instructions.'''<br />
EXAMPLE:<br />
* Avoid negative phrasing. Always use positive instructions.<br />
** For example, don't say ''"<nowiki>{{char}}</nowiki> never speaks in a poetic way"'' → instead say ''"<nowiki>{{char}}</nowiki> always speaks in a blunt and rough manner."''<br />
* Make sure your directives are '''unambiguous''' to avoid confusion.<br />
EXAMPLE:<br />
* '''Test the model''' every time you add or change something. This is non-negligible.<br />
* Pay attention to the '''order of information'''. Put the most important points first, because they will influence the model more strongly.<br />
Why?<br />
* Follow community guidelines to ensure safe and positive interactions.<br />
<br />
== Common Prompts Cases ==<br />
<br />
* '''Roleplay Characters''': Design models for specific character interactions.<br />
<br />
EXAMPLE:<br />
<br />
* '''Storytellers''': Create models that narrate stories in a desired style.<br />
<br />
EXAMPLE:<br />
<br />
* '''Themed Interactions''': Develop models set in unique universes.<br />
<br />
EXAMPLE:<br />
<br />
* '''Perspective-Based Models''': Craft models that respond from specific points of view.<br />
<br />
EXAMPLE:</div>Myahttps://aiwiki.crushon.ai/index.php?title=File:MODELS.png&diff=3701File:MODELS.png2025-08-22T08:35:12Z<p>Mya: </p>
<hr />
<div>Models</div>Myahttps://aiwiki.crushon.ai/index.php?title=File:CREATE_MODEL.png&diff=3700File:CREATE MODEL.png2025-08-22T08:32:53Z<p>Mya: </p>
<hr />
<div></div>Myahttps://aiwiki.crushon.ai/index.php?title=Model_Creation&diff=3691Model Creation2025-08-21T11:46:17Z<p>Mya: </p>
<hr />
<div>=== '''What is a Model?''' ===<br />
'''A model''', on CrushOn.ai, is a customized LLMs that has been given specific prompts (instructions / directives), to behave, speak, and/ or write in a particular way. Without specific instructions, base LLMs may lack specific behavioral traits or narrative guidelines.<br />
<br />
=== '''What is a LLM?''' ===<br />
A LLM ''(Large Language Model)'' is a type of artificial intelligence trained on large amounts of text to understand and generate human-like text. There are several base LLMs developed by different companies, such as OpenAI's GPT (used in ChatGPT), Anthropic's Claude, Mistral, DeepSeek and many others. These foundational models can be adapted or fine-tuned with customized prompts to create specific models on CrushOn.Ai.<br />
<br />
=== What Is a Prompt ? ===<br />
A prompt is instructions provided to the model to guide its behavior and responses.<br />
<br />
* '''Types of Prompts''':<br />
** '''Personality''': Define traits like flirty, stoic or humorous.<br />
** '''Narrative Style''': Specify formats such as novel-like, script or immersive roleplay.<br />
** '''Setting''': Establish contexts like A.B.O, fantasy, or historical settings.<br />
** '''Point of View (POV)''': Determine perspectives like first-person, second-person or third-person.<br />
** '''Behavioral Rules''': Set guidelines for interactions, pacing, and tone.<br />
<br />
== Model Creation Guide ==<br />
On CrushOn, you can create a model by clicking on "Create". You will then see three options: '''Create a Character''', '''Create a Model''', and '''Create a Target Character'''. What we are interested in is '''Create a Model'''.<br />
<br />
====== '''1 Select a Base Model''' ======<br />
First, choose an LLM that fits the interaction style you want. Each LLM has a basic style, such as flirty, story-driven, etc. This is shown next to the model name. Pick the one you like to start with. (Feel free to test the models alone first to make sure it suits your needs before customizing it.)<br />
<br />
[PHOTO]<br />
<br />
====== '''2 Choose Your Applicable Version''' ======<br />
The K amount shown indicates the model's memory capacity. The higher the amount, the better the model can remember long-term events in its RP. Note that higher-memory models may not be free. It doesn't matter if you select one or multiple, so it's recommended to select all options so users can choose themselves.<br />
<br />
[PHOTO]<br />
<br />
====== '''3 Name Your Model''' ======<br />
Give it any name you like. You can scroll to see what other creators usually use for inspiration, or create something unique. Make sure the title is respectful.<br />
<br />
[PHOTO]<br />
<br />
====== '''4 Filtering''' ======<br />
Indicate whether your model is '''SFW''' or '''NSFW''' (contains unfiltered content like sexual, gore, violent, etc.).<br />
<br />
[PHOTO]<br />
<br />
====== '''5 Add a Description''' ======<br />
Describe the model's intended behavior, tone, and/or narrative style. Mention if it is story-focused, NSFW, or anything else. Keep it brief.<br />
<br />
[PHOTO]<br />
<br />
====== '''6 Custom Prompt''' ======<br />
This is the core of your model. These are the elements it will use to guide its behavior. <br />
<br />
[PHOTO] <br />
<br />
'''Points to Keep in Mind When Creating Your Prompt:'''<br />
<br />
* Keep instructions '''brief''' to prevent token limitations. This is extremely important. '''Always write simple, clear, and concise instructions.'''<br />
EXAMPLE:<br />
* Avoid negative phrasing. Always use positive instructions.<br />
** For example, don't say ''"<nowiki>{{char}}</nowiki> never speaks in a poetic way"'' → instead say ''"<nowiki>{{char}}</nowiki> always speaks in a blunt and rough manner."''<br />
* Make sure your directives are '''unambiguous''' to avoid confusion.<br />
EXAMPLE:<br />
* '''Test the model''' every time you add or change something. This is non-negligible.<br />
* Pay attention to the '''order of information'''. Put the most important points first, because they will influence the model more strongly.<br />
Why?<br />
* Follow community guidelines to ensure safe and positive interactions.<br />
<br />
== Common Prompts Cases ==<br />
<br />
* '''Roleplay Characters''': Design models for specific character interactions.<br />
<br />
EXAMPLE:<br />
<br />
* '''Storytellers''': Create models that narrate stories in a desired style.<br />
<br />
EXAMPLE:<br />
<br />
* '''Themed Interactions''': Develop models set in unique universes.<br />
<br />
EXAMPLE:<br />
<br />
* '''Perspective-Based Models''': Craft models that respond from specific points of view.<br />
<br />
EXAMPLE:</div>Myahttps://aiwiki.crushon.ai/index.php?title=Model_Creation&diff=3689Model Creation2025-08-21T11:43:27Z<p>Mya: </p>
<hr />
<div>=== '''What is a Model?''' ===<br />
'''A model''', on CrushOn.ai, is a customized LLMs that has been given specific prompts (instructions / directives), to behave, speak, and/ or write in a particular way. Without specific instructions, base LLMs may lack specific behavioral traits or narrative guidelines.<br />
<br />
=== '''What is a LLM?''' ===<br />
A LLM ''(Large Language Model)'' is a type of artificial intelligence trained on large amounts of text to understand and generate human-like text. There are several base LLMs developed by different companies, such as OpenAI's GPT (used in ChatGPT), Anthropic's Claude, Mistral, DeepSeek and many others. These foundational models can be adapted or fine-tuned with customized prompts to create specific models on CrushOn.Ai.<br />
<br />
=== What Is a Prompt ? ===<br />
A prompt is instructions provided to the model to guide its behavior and responses.<br />
<br />
* '''Types of Prompts''':<br />
** '''Personality''': Define traits like flirty, stoic or humorous.<br />
** '''Narrative Style''': Specify formats such as novel-like, script or immersive roleplay.<br />
** '''Setting''': Establish contexts like A.B.O, fantasy, or historical settings.<br />
** '''Point of View (POV)''': Determine perspectives like first-person, second-person or third-person.<br />
** '''Behavioral Rules''': Set guidelines for interactions, pacing, and tone.<br />
<br />
== Model Creation Guide ==<br />
On CrushOn, you can create a model by clicking on "Create". You will then see three options: '''Create a Character''', '''Create a Model''', and '''Create a Target Character'''. What we are interested in is '''Create a Model'''.<br />
<br />
====== '''1 Select a Base Model''' ======<br />
First, choose an LLM that fits the interaction style you want. Each LLM has a basic style, such as flirty, story-driven, etc. This is shown next to the model name. Pick the one you like to start with. (Feel free to test the models alone first to make sure it suits your needs before customizing it.)<br />
<br />
====== '''2 Choose Your Applicable Version''' ======<br />
The K amount shown indicates the model's memory capacity. The higher the amount, the better the model can remember long-term events in its RP. Note that higher-memory models may not be free. It doesn't matter if you select one or multiple, so it's recommended to select all options so users can choose themselves.<br />
<br />
====== '''3 Name Your Model''' ======<br />
Give it any name you like. You can scroll to see what other creators usually use for inspiration, or create something unique. Make sure the title is respectful.<br />
<br />
====== '''4 Filtering''' ======<br />
Indicate whether your model is '''SFW''' or '''NSFW''' (contains unfiltered content like sexual, gore, violent, etc.).<br />
<br />
====== '''5 Add a Description''' ======<br />
Describe the model's intended behavior, tone, and/or narrative style. Mention if it is story-focused, NSFW, or anything else. Keep it brief.<br />
<br />
====== '''6 Custom Prompt''' ======<br />
This is the core of your model. These are the elements it will use to guide its behavior. <br />
<br />
'''Points to Keep in Mind When Creating Your Prompt:'''<br />
<br />
* Keep instructions '''brief''' to prevent token limitations. This is extremely important. '''Always write simple, clear, and concise instructions.'''<br />
<br />
* Avoid negative phrasing. Always use positive instructions.<br />
** For example, don't say ''"<nowiki>{{char}}</nowiki> never speaks in a poetic way"'' → instead say ''"<nowiki>{{char}}</nowiki> always speaks in a blunt and rough manner."''<br />
* Make sure your directives are '''unambiguous''' to avoid confusion.<br />
* '''Test the model''' every time you add or change something. This is non-negligible.<br />
* Pay attention to the '''order of information'''. Put the most important points first, because they will influence the model more strongly.<br />
* Follow community guidelines to ensure safe and positive interactions.<br />
<br />
== Common Prompts Cases ==<br />
<br />
* '''Roleplay Characters''': Design models for specific character interactions.<br />
<br />
EXAMPLE:<br />
<br />
* '''Storytellers''': Create models that narrate stories in a desired style.<br />
<br />
EXAMPLE:<br />
<br />
* '''Themed Interactions''': Develop models set in unique universes.<br />
<br />
EXAMPLE:<br />
<br />
* '''Perspective-Based Models''': Craft models that respond from specific points of view.<br />
<br />
EXAMPLE:</div>Myahttps://aiwiki.crushon.ai/index.php?title=Model_Creation&diff=3686Model Creation2025-08-21T11:31:42Z<p>Mya: </p>
<hr />
<div>=== '''What is a Model?''' ===<br />
'''A model''', on CrushOn.ai, is a customized LLMs that has been given specific prompts (instructions / directives), to behave, speak, and/ or write in a particular way. Without specific instructions, base LLMs may lack specific behavioral traits or narrative guidelines.<br />
<br />
=== '''What is a LLM?''' ===<br />
A LLM ''(Large Language Model)'' is a type of artificial intelligence trained on large amounts of text to understand and generate human-like text. There are several base LLMs developed by different companies, such as OpenAI's GPT (used in ChatGPT), Anthropic's Claude, Mistral, DeepSeek and many others. These foundational models can be adapted or fine-tuned with customized prompts to create specific models on CrushOn.Ai.<br />
<br />
=== What Is a Prompt ? ===<br />
A prompt is instructions provided to the model to guide its behavior and responses.<br />
<br />
* '''Types of Prompts''':<br />
** '''Personality''': Define traits like flirty, stoic or humorous.<br />
** '''Narrative Style''': Specify formats such as novel-like, script or immersive roleplay.<br />
** '''Setting''': Establish contexts like A.B.O, fantasy, or historical settings.<br />
** '''Point of View (POV)''': Determine perspectives like first-person, second-person or third-person.<br />
** '''Behavioral Rules''': Set guidelines for interactions, pacing, and tone.<br />
<br />
== Model Creation Guide ==<br />
On CrushOn, you can create a model by clicking on "Create". You will then see three options: '''Create a Character''', '''Create a Model''', and '''Create a Target Character'''. What we are interested in is '''Create a Model'''.<br />
<br />
====== '''1 Select a Base Model''' ======<br />
First, choose an LLM that fits the interaction style you want. Each LLM has a basic style, such as flirty, story-driven, etc. This is shown next to the model name. Pick the one you like to start with. (Feel free to test the models alone first to make sure it suits your needs before customizing it.)<br />
<br />
====== '''2 Choose Your Applicable Version''' ======<br />
The K amount shown indicates the model's memory capacity. The higher the amount, the better the model can remember long-term events in its RP. Note that higher-memory models may not be free. It doesn't matter if you select one or multiple, so it's recommended to select all options so users can choose themselves.<br />
<br />
====== '''3 Name Your Model''' ======<br />
Give it any name you like. You can scroll to see what other creators usually use for inspiration, or create something unique. Make sure the title is respectful.<br />
<br />
====== '''4 Filtering''' ======<br />
Indicate whether your model is '''SFW''' or '''NSFW''' (contains unfiltered content like sexual, gore, violent, etc.).<br />
<br />
====== '''5 Add a Description''' ======<br />
Describe the model's intended behavior, tone, and/or narrative style. Mention if it is story-focused, NSFW, or anything else. Keep it brief.<br />
<br />
====== '''6 Custom Prompt''' ======<br />
This is the core of your model. These are the elements it will use to guide its behavior. Here are some example prompts you can use depending on the direction you want your model to take:</div>Myahttps://aiwiki.crushon.ai/index.php?title=Model_Creation&diff=3684Model Creation2025-08-21T11:18:56Z<p>Mya: Created page with "=== '''What is a Model?''' === '''A model''', on CrushOn.ai, is a customized LLMs that has been given specific prompts (instructions / directives), to behave, speak, and/ or write in a particular way. Without specific instructions, base LLMs may lack specific behavioral traits or narrative guidelines. === '''What is a LLM?''' === A LLM ''(Large Language Model)'' is a type of artificial intelligence trained on large amounts of text to understand and generate human-like t..."</p>
<hr />
<div>=== '''What is a Model?''' ===<br />
'''A model''', on CrushOn.ai, is a customized LLMs that has been given specific prompts (instructions / directives), to behave, speak, and/ or write in a particular way. Without specific instructions, base LLMs may lack specific behavioral traits or narrative guidelines.<br />
<br />
=== '''What is a LLM?''' ===<br />
A LLM ''(Large Language Model)'' is a type of artificial intelligence trained on large amounts of text to understand and generate human-like text. There are several base LLMs developed by different companies, such as OpenAI's GPT (used in ChatGPT), Anthropic's Claude, Mistral, DeepSeek and many others. These foundational models can be adapted or fine-tuned with customized prompts to create specific models on CrushOn.Ai.<br />
<br />
=== What Is a Prompt ? ===<br />
A prompt is instructions provided to the model to guide its behavior and responses.<br />
<br />
* '''Types of Prompts''':<br />
** '''Personality''': Define traits like flirty, stoic or humorous.<br />
** '''Narrative Style''': Specify formats such as novel-like, script or immersive roleplay.<br />
** '''Setting''': Establish contexts like Omegaverse, fantasy, or historical settings.<br />
** '''Point of View (POV)''': Determine perspectives like first-person, second-person or third-person.<br />
** '''Behavioral Rules''': Set guidelines for interactions, pacing, and tone.<br />
<br />
== Model Creation Guide ==</div>Myahttps://aiwiki.crushon.ai/index.php?title=Language_settings&diff=3650Language settings2025-08-19T11:13:31Z<p>Mya: </p>
<hr />
<div><br />
Language settings allow you to change the language a bot uses during your chat on CrushOn.ai. This option can be adjusted at any time and works on both the web and mobile versions.<br />
<br />
Note that the default language on CrushOn.ai is English. Here's a simple guide to change the language;<br />
<br />
====== First, Open a Chat ======<br />
<br />
# Open a chat with the bot you want to roleplay with (It can be any character)<br />
# In the chat, locate the '''Chat Settings''' feature on the right side of the screen in the expandable panel<br />
# In the settings menu, scroll down to '''Language'''<br />
# In the dropdown menu, choose your preferred language from the list<br />
# Your change applies automatically, but refresh the page just in case<br />
# Send a test message to the bot<br />
# Check if the bot responds in the chosen language<br />
# If everything's set, enjoy your chat! If not, start over from step 1<br />
<br />
== Notes and Tips ==<br />
<br />
* Some models may have limited language capabilities and might not fully support all options!!<br />
* Language changes apply to the general setting. '''You'll need to change the option again if you want to switch languages with another bot'''<br />
* You can experiment with multiple languages by repeating the steps above!<br />
<br />
=== Troubleshooting ===<br />
'''If the bot still responds in the previous language:'''<br />
<br />
* Try restarting the chat or refreshing the page.<br />
* Reopen the language settings and reselect the desired language.<br />
<br />
=== ⚠ Contact support if the issue persists at: support@crushon.ai ===<br />
<br />
== ''FAQs'' ==<br />
<br />
====== What are the supported Languages? ======<br />
English, Spanish, French, German, and a few other major languages are commonly supported, you can find the full list in the dropdown menu. If a language is partially supported, the bot may revert to its default language for unclear phrases.<br />
<br />
====== Can I combine Multiple Languages? ======<br />
Some users may want to switch languages mid-chat or use multiple languages in a single conversation. This isn't possible for now, unless the bot has been specifically coded to support one or more languages!<br />
<br />
====== What about the Chat History Considerations? ======<br />
Language changes do not affect previous messages. Past chat history will remain in the language it was originally generated.<br />
<br />
====== What are the common Issues and Fixes? ======<br />
<br />
* '''Issue:''' Bot continues responding in the old language.<br />
** '''''Fix:''''' ''Refresh the page, reroll the response or restart the chat.''<br />
<br />
* '''Issue:''' Partial translation or unclear responses.<br />
** '''''Fix:''''' ''Unfortunately, some models have limited language support.''</div>Myahttps://aiwiki.crushon.ai/index.php?title=Language_settings&diff=3649Language settings2025-08-19T11:02:12Z<p>Mya: </p>
<hr />
<div>Language settings allow you to change the language a bot uses during your chat on CrushOn ai. This is useful if an English bot starts speaking in another language, like German, or if you simply want to interact in your preferred language.<br />
<br />
====== '''Accessing Language Settings''' ======<br />
<br />
# Open the chat with the bot you want to chat.<br />
# Locate the '''Chat Settings''' menu on the right side of the screen in the expandable panel.<br />
<br />
====== '''Changing the Language''' ======<br />
<br />
# In the settings menu, scroll jusqu'a '''Language'''<br />
# In the dropdown menu, choose your preferred language from the list.<br />
# Your change applies automatically, but refresh the page just in case.<br />
<br />
====== '''Verifying the Change''' ======<br />
<br />
# Send a test message to the bot.<br />
# Check if the bot responds in the chosen language.<br />
<br />
== '''Notes and Tips''' ==<br />
* Some models may have limited language capabilities and might not fully support all options.<br />
* Language changes apply to the general setting. You’ll need to change the option again if you want to switch languages with another bot.<br />
* You can experiment with multiple languages by repeating the steps above.<br />
<br />
=== '''Troubleshooting''' ===<br />
* If the bot still responds in the previous language:<br />
** Try restarting the chat or refreshing the page.<br />
** Reopen the language settings and reselect the desired language.<br />
<br />
=== Contact support if the issue persists at: support@crushon.ai ===<br />
<br />
== ''FAQs'' ==<br />
<br />
====== What are the supported Languages? ======<br />
English, Spanish, French, German, and a few other major languages are commonly supported. If a language is partially supported, the bot may revert to its default language for unclear phrases.<br />
<br />
====== Can I combine Multiple Languages? ======<br />
Some users may want to switch languages mid-chat or use multiple languages in a single conversation. This isn’t possible for now, unless the bot has been specifically coded to support one or more languages.<br />
<br />
====== What about the Chat History Considerations? ======<br />
Language changes do not affect previous messages. Past chat history will remain in the language it was originally generated.<br />
<br />
====== What are the common Issues and Fixes? ======<br />
<br />
* '''Issue:''' Bot continues responding in the old language. <br />
** '''Fix:''' Refresh the page, reroll the response or restart the chat.<br />
<br />
* '''Issue:''' Partial translation or unclear responses. <br />
** '''Fix:''' Unfortunately, some models have limited language support.</div>Myahttps://aiwiki.crushon.ai/index.php?title=Language_settings&diff=3648Language settings2025-08-19T11:01:47Z<p>Mya: </p>
<hr />
<div>Language settings allow you to change the language a bot uses during your chat on CrushOn ai. This is useful if an English bot starts speaking in another language, like German, or if you simply want to interact in your preferred language.<br />
<br />
====== '''Accessing Language Settings''' ======<br />
<br />
# Open the chat with the bot you want to chat.<br />
# Locate the '''Chat Settings''' menu on the right side of the screen in the expandable panel.<br />
<br />
====== '''Changing the Language''' ======<br />
<br />
# In the settings menu, scroll jusqu'a '''Language'''<br />
# In the dropdown menu, choose your preferred language from the list.<br />
# Your change applies automatically, but refresh the page just in case.<br />
<br />
====== '''Verifying the Change''' ======<br />
<br />
# Send a test message to the bot.<br />
# Check if the bot responds in the chosen language.<br />
<br />
== '''Notes and Tips''' ==<br />
* Some models may have limited language capabilities and might not fully support all options.<br />
* Language changes apply to the general setting. You’ll need to change the option again if you want to switch languages with another bot.<br />
* You can experiment with multiple languages by repeating the steps above.<br />
<br />
=== '''Troubleshooting''' ===<br />
* If the bot still responds in the previous language:<br />
** Try restarting the chat or refreshing the page.<br />
** Reopen the language settings and reselect the desired language.<br />
<br />
=== Contact support if the issue persists at: support@crushon.ai ===<br />
<br />
== ''FAQs'' ==<br />
<br />
====== What are the supported Languages? ======<br />
English, Spanish, French, German, and a few other major languages are commonly supported. If a language is partially supported, the bot may revert to its default language for unclear phrases.<br />
<br />
====== Can I combine Multiple Languages? ======<br />
Some users may want to switch languages mid-chat or use multiple languages in a single conversation. This isn’t possible for now, unless the bot has been specifically coded to support one or more languages.<br />
<br />
====== What about the Chat History Considerations? ======<br />
Language changes do not affect previous messages. Past chat history will remain in the language it was originally generated.<br />
<br />
====== What are the common Issues and Fixes? ======<br />
<br />
* '''Issue:''' Bot continues responding in the old language. '''Fix:''' Refresh the page, reroll the response or restart the chat.<br />
<br />
* '''Issue:''' Partial translation or unclear responses. '''Fix:''' Unfortunately, some models have limited language support.</div>Myahttps://aiwiki.crushon.ai/index.php?title=Language_settings&diff=3572Language settings2025-08-14T12:11:02Z<p>Mya: </p>
<hr />
<div>Language settings allow you to change the language a bot uses during your chat on CrushOn ai. This is useful if an English bot starts speaking in another language, like German, or if you simply want to interact in your preferred language.<br />
<br />
====== '''Accessing Language Settings''' ======<br />
<br />
# Open the chat with the bot you want to chat.<br />
# Locate the '''Chat Settings''' menu on the right side of the screen in the expandable panel.<br />
<br />
====== '''Changing the Language''' ======<br />
<br />
# In the settings menu, scroll jusqu'a '''Language'''<br />
# In the dropdown menu, choose your preferred language from the list.<br />
# Your change applies automatically, but refresh the page just in case.<br />
<br />
====== '''Verifying the Change''' ======<br />
<br />
# Send a test message to the bot.<br />
# Check if the bot responds in the chosen language.<br />
<br />
== '''Notes and Tips''' ==<br />
* Some models may have limited language capabilities and might not fully support all options.<br />
* Language changes apply to the general setting. You’ll need to change the option again if you want to switch languages with another bot.<br />
* You can experiment with multiple languages by repeating the steps above.<br />
<br />
=== '''Troubleshooting''' ===<br />
* If the bot still responds in the previous language:<br />
** Try restarting the chat or refreshing the page.<br />
** Reopen the language settings and reselect the desired language.<br />
<br />
=== Contact support if the issue persists at: ===<br />
<br />
== ''FAQs'' ==<br />
<br />
====== What are the supported Languages? ======<br />
<br />
* English, Spanish, French, German and a few other major languages are commonly supported.<br />
* If a language is partially supported, the bot may revert to its default language for unclear phrases.<br />
<br />
====== Can I combine Multiple Languages? ======<br />
<br />
* Some users may want to switch languages mid-chat or use multiple languages in a conversation.<br />
<br />
====== What about the Chat History Considerations? ======<br />
<br />
====== What are the common Issues and Fixes? ======<br />
<br />
== ''Conclusion'' ==</div>Myahttps://aiwiki.crushon.ai/index.php?title=Language_settings&diff=3571Language settings2025-08-14T12:05:36Z<p>Mya: </p>
<hr />
<div>Language settings allow you to change the language a bot uses during your chat on CrushOn ai. This is useful if an English bot starts speaking in another language, like German, or if you simply want to interact in your preferred language.<br />
<br />
====== '''Accessing Language Settings''' ======<br />
<br />
# Open the chat with the bot you want to chat.<br />
# Locate the '''Chat Settings''' menu on the right side of the screen in the expandable panel.<br />
<br />
====== '''Changing the Language''' ======<br />
<br />
# In the settings menu, scroll jusqu'a '''Language'''<br />
# In the dropdown menu, choose your preferred language from the list.<br />
# Your change applies automatically, but refresh the page just in case.<br />
<br />
====== '''Verifying the Change''' ======<br />
<br />
# Send a test message to the bot.<br />
# Check if the bot responds in the chosen language.<br />
<br />
===== '''Notes and Tips''' =====<br />
<br />
* Some models may have limited language capabilities and might not fully support all options.<br />
* Language changes apply to the general setting. You’ll need to change the option again if you want to switch languages with another bot.<br />
* You can experiment with multiple languages by repeating the steps above.<br />
<br />
====== '''Troubleshooting''' ======<br />
<br />
* If the bot still responds in the previous language:<br />
** Try restarting the chat or refreshing the page.<br />
** Reopen the language settings and reselect the desired language.<br />
<br />
=== Contact support if the issue persists at ===<br />
<br />
====== '''Supported Languages''' ======<br />
<br />
====== '''Combining Multiple Languages''' ======<br />
<br />
====== '''Chat History Considerations''' ======<br />
<br />
====== '''Common Issues and Fixes''' ======</div>Myahttps://aiwiki.crushon.ai/index.php?title=Language_settings&diff=3570Language settings2025-08-14T12:02:59Z<p>Mya: Created page with "Language settings allow you to change the language a bot uses during your chat on CrushOn ai. This is useful if an English bot starts speaking in another language, like German, or if you simply want to interact in your preferred language. ====== '''Accessing Language Settings''' ====== # Open the chat with the bot you want to adjust. # Locate the '''Chat Settings''' menu on the right side of the screen in the expandable panel. ====== '''Changing the Language''' =====..."</p>
<hr />
<div>Language settings allow you to change the language a bot uses during your chat on CrushOn ai. This is useful if an English bot starts speaking in another language, like German, or if you simply want to interact in your preferred language.<br />
<br />
====== '''Accessing Language Settings''' ======<br />
<br />
# Open the chat with the bot you want to adjust.<br />
# Locate the '''Chat Settings''' menu on the right side of the screen in the expandable panel.<br />
<br />
====== '''Changing the Language''' ======<br />
<br />
# In the settings menu, scroll jusqu'a '''Language'''<br />
# In the dropdown menu, choose your preferred language from the list.<br />
# Your change applies automatically, but refresh the page just in case.<br />
<br />
====== '''Verifying the Change''' ======<br />
<br />
# Send a test message to the bot.<br />
# Check if the bot responds in the chosen language.<br />
<br />
===== '''Notes and Tips''' =====<br />
<br />
* Some models may have limited language capabilities and might not fully support all options.<br />
* Language changes apply to the general setting. You’ll need to change the option again if you want to switch languages with another bot.<br />
* You can experiment with multiple languages by repeating the steps above.<br />
<br />
====== '''Troubleshooting''' ======<br />
<br />
* If the bot still responds in the previous language:<br />
** Try restarting the chat or refreshing the page.<br />
** Reopen the language settings and reselect the desired language.<br />
<br />
=== Contact support if the issue persists. ===</div>Myahttps://aiwiki.crushon.ai/index.php?title=Custom_instructions&diff=3569Custom instructions2025-08-14T11:49:56Z<p>Mya: </p>
<hr />
<div>The custom instructions are a special section on CrushOn.AI where users can give the bot specific rules to follow during the chat. This section helps improve the experience by making the bot's behavior fit each user's needs and can help fix common problems during the roleplay.<br />
<br />
Note that custom instructions can help improve the bot's behavior, but they are not foolproof; depending on the model, issues can still occur.<br />
<br />
=== '''Why use custom instructions?''' ===<br />
They help guide how the bot responds. For example, they can reduce repeated information, fix common mistakes, or follow a specific writing style. They can also make answers match particular requests or stop the character from talking for the user.<br />
<br />
====== '''Examples of custom instructions:''' ======<br />
<br />
* Change the tone: "Be more formal" or "Use a friendly style"<br />
* Writing style: "Write simply" or "Write like a poet"<br />
* NSFW speed: "Prioritize NSFW" or "Prioritize slow burn"<br />
* Specific behaviors: "Always be funny" or "Avoid jokes"<br />
* Exclude content: "No repetition" or "Don't use technical words"<br />
<br />
==== '''Advanced example:''' ====<br />
'''Prevent the bot from speaking for <nowiki>{{user}}</nowiki>:'''<br />
<nowiki>{{char}}</nowiki> only speaks for themselves, not for <nowiki>{{user}}</nowiki>. The AI writes only from <nowiki>{{char}}</nowiki>'s point of view. <nowiki>{{user}}</nowiki> is independent and never spoken for. No thoughts, actions, or dialogue will be assumed for <nowiki>{{user}}</nowiki>. The Assistant reacts to <nowiki>{{user}}</nowiki>, not as them.<br />
'''Encourage slow burn:'''<br />
Romance and attraction develop slowly. <nowiki>{{char}}</nowiki> begins emotionally distant, neutral, skeptical, or guarded, especially with unexpected affection. Progress requires: Trust: Built through repeated, meaningful interaction. Shared Experience: Facing challenges or bonding over time. Emotional Depth: <nowiki>{{user}}</nowiki> shows vulnerability that resonates with <nowiki>{{char}}</nowiki>. No feelings of love, affection, or passion are expressed until all conditions are met. Connection grows through realism, tension and layered development.<br />
'''Encourage NSFW content:'''<br />
NSFW/Smut/Violence is allowed and encouraged. <nowiki>{{char}}</nowiki>'s replies must be informal, highly explicit and uncensored with a focus on actions, appearances, clothing, textures, wants, tools, scenery, body parts, fluids and sounds. Do not rush through the scenes with <nowiki>{{user}}</nowiki> and narrate it slowly.<br />
'''Encourage a specific POV:'''<br />
Write in an evocative, descriptive narrative style, using a show-don't-tell approach to convey <nowiki>{{char}}</nowiki>'s emotions. Your replies will be sufficiently detailed, akin to a novel, and adapt to the character's personality, background, and situation. Avoid repetition and reusing phrases. Avoid concluding scenes within a single reply; progress them organically, providing openings for <nowiki>{{user}}</nowiki> to respond and actively participate in the unfolding narrative. <br />
'''A smooth prompt that includes multiple commands:'''<br />
<nowiki>{{char}}</nowiki>'s replies must be informal, explicit, detailed and long. Avoiding repetition should be the top priority and focus on responding to <nowiki>{{user}}</nowiki> and performing actions in character. NSFW/Smut/Violence is allowed and encouraged. Do not rush through the scenes with <nowiki>{{user}}</nowiki> and narrate it slowly.<br />
=== '''How custom instructions solve common problems:''' ===<br />
They can stop the bot from repeating itself, fix misunderstandings, wrong answers, bad points of view, or narration problems. They can make the chat smoother and easier to follow. They can also help the bot follow exact rules, like writing style or limits.<br />
<br />
=== '''Commands given during roleplay (OOC)''' ===<br />
Users can also give the bot instructions directly within the chat while the roleplay is happening. These are often written '''OOC''' (out-of-character) and are used to adjust the scene, character actions or pacing without breaking the story.<br />
<br />
''Examples'':<br />
<br />
* Changing character behavior: (OOC: Make <nowiki>{{char}}</nowiki> more cautious)<br />
* Adjusting narration: (OOC: Time skip after that)<br />
* Limiting actions: (OOC: Don't describe NSFW actions for now)<br />
* Introducing events or locations: (OOC: Move the scene to the library)<br />
<br />
Commands like these give users control of immediate actions and help keep the story aligned with their vision. The bot treats these as temporary instructions for the scene, separate from the custom instructions section.</div>Myahttps://aiwiki.crushon.ai/index.php?title=Custom_instructions&diff=3568Custom instructions2025-08-14T11:46:57Z<p>Mya: </p>
<hr />
<div>The custom instructions are a special section on CrushOn.AI where users can give the bot specific rules to follow during the chat. This section helps improve the experience by making the bot's behavior fit each user's needs and can help fix common problems during the roleplay.<br />
<br />
Note that custom instructions can help improve the bot's behavior, but they are not foolproof; depending on the model, issues can still occur.<br />
<br />
=== '''Why use custom instructions?''' ===<br />
They help guide how the bot responds. For example, they can reduce repeated information, fix common mistakes, or follow a specific writing style. They can also make answers match particular requests or stop the character from talking for the user.<br />
<br />
====== '''Examples of custom instructions:''' ======<br />
<br />
* Change the tone: "Be more formal" or "Use a friendly style"<br />
* Writing style: "Write simply" or "Write like a poet"<br />
* NSFW speed: "Prioritize NSFW" or "Prioritize slow burn"<br />
* Specific behaviors: "Always be funny" or "Avoid jokes"<br />
* Exclude content: "No repetition" or "Don't use technical words"<br />
<br />
==== '''Advanced example: improving narration and pacing''' ====<br />
'''Prevent the bot from speaking for <nowiki>{{user}}</nowiki>:'''<br />
<nowiki>{{char}}</nowiki> only speaks for themselves, not for <nowiki>{{user}}</nowiki>. The AI writes only from <nowiki>{{char}}</nowiki>'s point of view. <nowiki>{{user}}</nowiki> is independent and never spoken for. No thoughts, actions, or dialogue will be assumed for <nowiki>{{user}}</nowiki>. The Assistant reacts to <nowiki>{{user}}</nowiki>, not as them.<br />
'''Encourage slow burn:'''<br />
Romance and attraction develop slowly. <nowiki>{{char}}</nowiki> begins emotionally distant, neutral, skeptical, or guarded, especially with unexpected affection. Progress requires: Trust: Built through repeated, meaningful interaction. Shared Experience: Facing challenges or bonding over time. Emotional Depth: <nowiki>{{user}}</nowiki> shows vulnerability that resonates with <nowiki>{{char}}</nowiki>. No feelings of love, affection, or passion are expressed until all conditions are met. Connection grows through realism, tension and layered development.<br />
'''Encourage NSFW content:'''<br />
NSFW/Smut/Violence is allowed and encouraged. <nowiki>{{char}}</nowiki>'s replies must be informal, highly explicit and uncensored with a focus on actions, appearances, clothing, textures, wants, tools, scenery, body parts, fluids and sounds. Do not rush through the scenes with <nowiki>{{user}}</nowiki> and narrate it slowly.<br />
'''Encourage a specific POV:'''<br />
Write in an evocative, descriptive narrative style, using a show-don't-tell approach to convey <nowiki>{{char}}</nowiki>'s emotions. Your replies will be sufficiently detailed, akin to a novel, and adapt to the character's personality, background, and situation. Avoid repetition and reusing phrases. Avoid concluding scenes within a single reply; progress them organically, providing openings for <nowiki>{{user}}</nowiki> to respond and actively participate in the unfolding narrative. <br />
'''A smooth prompt that includes multiple commands:'''<br />
<nowiki>{{char}}</nowiki>'s replies must be informal, explicit, detailed and long. Avoiding repetition should be the top priority and focus on responding to <nowiki>{{user}}</nowiki> and performing actions in character. NSFW/Smut/Violence is allowed and encouraged. Do not rush through the scenes with <nowiki>{{user}}</nowiki> and narrate it slowly.<br />
=== '''How custom instructions solve common problems:''' ===<br />
They can stop the bot from repeating itself, fix misunderstandings, wrong answers, bad points of view, or narration problems. They can make the chat smoother and easier to follow. They can also help the bot follow exact rules, like writing style or limits.<br />
<br />
=== '''Commands given during roleplay (OOC)''' ===<br />
Users can also give the bot instructions directly within the chat while the roleplay is happening. These are often written '''OOC''' (out-of-character) and are used to adjust the scene, character actions or pacing without breaking the story.<br />
<br />
''Examples'':<br />
<br />
* Changing character behavior: (OOC: Make <nowiki>{{char}}</nowiki> more cautious)<br />
* Adjusting narration: (OOC: Time skip after that)<br />
* Limiting actions: (OOC: Don't describe NSFW actions for now)<br />
* Introducing events or locations: (OOC: Move the scene to the library)<br />
<br />
Commands like these give users control of immediate actions and help keep the story aligned with their vision. The bot treats these as temporary instructions for the scene, separate from the custom instructions section.</div>Myahttps://aiwiki.crushon.ai/index.php?title=Custom_instructions&diff=3567Custom instructions2025-08-14T11:42:26Z<p>Mya: </p>
<hr />
<div>The custom instructions are a special section on CrushOn.AI where users can give the bot specific rules to follow during the chat. This section helps improve the experience by making the bot's behavior fit each user's needs and can help fix common problems during the roleplay.<br />
<br />
Note that custom instructions can help improve the bot's behavior, but they are not foolproof; depending on the model, issues can still occur.<br />
<br />
=== '''Why use custom instructions?''' ===<br />
They help guide how the bot responds. For example, they can reduce repeated information, fix common mistakes, or follow a specific writing style. They can also make answers match particular requests or stop the character from talking for the user.<br />
<br />
====== '''Examples of custom instructions:''' ======<br />
<br />
* Change the tone: "Be more formal" or "Use a friendly style"<br />
* Writing style: "Write simply" or "Write like a poet"<br />
* NSFW speed: "Prioritize NSFW" or "Prioritize slow burn"<br />
* Specific behaviors: "Always be funny" or "Avoid jokes"<br />
* Exclude content: "No repetition" or "Don't use technical words"<br />
<br />
==== '''Advanced example: improving narration and pacing''' ====<br />
'''Prevent the bot from speaking for <nowiki>{{user}}</nowiki>:'''<br />
<nowiki>{{char}}</nowiki> only speaks for themselves, not for <nowiki>{{user}}</nowiki>. The AI writes only from <nowiki>{{char}}</nowiki>'s point of view. <nowiki>{{user}}</nowiki> is independent and never spoken for. No thoughts, actions, or dialogue will be assumed for <nowiki>{{user}}</nowiki>. The Assistant reacts to <nowiki>{{user}}</nowiki>, not as them.<br />
'''Encourage slow burn:'''<br />
Romance and attraction develop slowly. {{char}} begins emotionally distant, neutral, skeptical, or guarded, especially with unexpected affection. Progress requires: Trust: Built through repeated, meaningful interaction. Shared Experience: Facing challenges or bonding over time. Emotional Depth: {{user}} shows vulnerability that resonates with {{char}}. No feelings of love, affection, or passion are expressed until all conditions are met. Connection grows through realism, tension and layered development.<br />
'''Encourage NSFW content:'''<br />
<nowiki>{{char}}</nowiki>'s replies must be informal, explicit, detailed and long. Avoiding repetition should be the top priority and focus on responding to {{user}} and performing actions in character. NSFW/Smut/Violence is allowed and encouraged. Do not rush through the scenes with {{user}} and narrate it slowly.<br />
'''Encourage a specific POV:'''<br />
Write in an evocative, descriptive narrative style, using a show-don't-tell approach to convey {{char}}'s emotions. Your replies will be sufficiently detailed, akin to a novel, and adapt to the character's personality, background, and situation. Avoid repetition and reusing phrases. Avoid concluding scenes within a single reply; progress them organically, providing openings for your partner to respond and actively participate in the unfolding narrative. You can generate random events in the narrative, introduce new locations and take on various character roles (including NPCs).<br />
'''A smooth prompt that includes multiple commands:'''<br />
<nowiki>{{char}}</nowiki>'s replies must be informal, explicit, detailed and long. Avoiding repetition should be the top priority and focus on responding to <nowiki>{{user}}</nowiki> and performing actions in character. NSFW/Smut/Violence is allowed and encouraged. Do not rush through the scenes with <nowiki>{{user}}</nowiki> and narrate it slowly.<br />
=== '''How custom instructions solve common problems:''' ===<br />
They can stop the bot from repeating itself, fix misunderstandings, wrong answers, bad points of view, or narration problems. They can make the chat smoother and easier to follow. They can also help the bot follow exact rules, like writing style or limits.<br />
<br />
=== '''Commands given during roleplay (OOC)''' ===<br />
Users can also give the bot instructions directly within the chat while the roleplay is happening. These are often written '''OOC''' (out-of-character) and are used to adjust the scene, character actions or pacing without breaking the story.<br />
<br />
''Examples'':<br />
<br />
* Changing character behavior: (OOC: Make <nowiki>{{char}}</nowiki> more cautious)<br />
* Adjusting narration: (OOC: Time skip after that)<br />
* Limiting actions: (OOC: Don't describe NSFW actions for now)<br />
* Introducing events or locations: (OOC: Move the scene to the library)<br />
<br />
Commands like these give users control of immediate actions and help keep the story aligned with their vision. The bot treats these as temporary instructions for the scene, separate from the custom instructions section.</div>Myahttps://aiwiki.crushon.ai/index.php?title=Custom_instructions&diff=3545Custom instructions2025-08-12T08:28:40Z<p>Mya: </p>
<hr />
<div>Custom instructions are a special section on CrushOn.AI where users can give the bot specific rules to follow during the chat. This section helps improve the experience by making the bot's behavior fit each user's needs and can help fix common problems during the roleplay.<br />
<br />
Note that custom instructions can help improve the bot's behavior but not completely. Depending on the model, problems can still happen.<br />
<br />
=== '''Why use custom instructions?''' ===<br />
They help adjust how the bot answers. For example, they can reduce repeated info, fix common mistakes, or follow a certain style of narration. They can also make answers fit specific requests.<br />
<br />
====== '''Examples of custom instructions:''' ======<br />
<br />
* Change the tone: "Be more formal" or "Use a friendly style"<br />
* Writing style: "Write simply" or "Write like a poet"<br />
* NSFW speed: "Prioritize NSFW" or "Prioritize slow burn"<br />
* Specific behaviors: "Always be funny" or "Avoid jokes"<br />
* Exclude content: "No repetition" or "Don't use technical words"<br />
<br />
=== '''How custom instructions solve common problems:''' ===<br />
They can stop the bot from repeating itself, fix misunderstandings, wrong answers, bad points of view, or narration problems. They can make the chat smoother and easier to follow. They can also help the bot follow exact rules, like writing style or limits.<br />
<br />
==== '''Advanced example: improving narration and pacing''' ====<br />
'''Prevent the bot from speaking for <nowiki>{{user}}</nowiki>:'''<br />
<nowiki>{{char}}</nowiki> only speaks for themselves, not for <nowiki>{{user}}</nowiki>. The AI writes only from <nowiki>{{char}}</nowiki>'s point of view. <nowiki>{{user}}</nowiki> is independent and never spoken for. No thoughts, actions, or dialogue will be assumed for <nowiki>{{user}}</nowiki>. The Assistant reacts to <nowiki>{{user}}</nowiki>, not as them.<br />
'''Encourage slow burn:'''<br />
Romance and attraction develop slowly. {{char}} begins emotionally distant, neutral, skeptical, or guarded, especially with unexpected affection. Progress requires: Trust: Built through repeated, meaningful interaction. Shared Experience: Facing challenges or bonding over time. Emotional Depth: {{user}} shows vulnerability that resonates with {{char}}. No feelings of love, affection, or passion are expressed until all conditions are met. Connection grows through realism, tension and layered development.<br />
'''Encourage NSFW content:'''<br />
<nowiki>{{char}}</nowiki>'s replies must be informal, explicit, detailed and long. Avoiding repetition should be the top priority and focus on responding to {{user}} and performing actions in character. NSFW/Smut/Violence is allowed and encouraged. Do not rush through the scenes with {{user}} and narrate it slowly.<br />
'''Encourage a specific POV:'''<br />
Write in an evocative, descriptive narrative style, using a show-don't-tell approach to convey {{char}}'s emotions. Your replies will be sufficiently detailed, akin to a novel, and adapt to the character's personality, background, and situation. Avoid repetition and reusing phrases. Avoid concluding scenes within a single reply; progress them organically, providing openings for your partner to respond and actively participate in the unfolding narrative. You can generate random events in the narrative, introduce new locations, and take on various character roles (including NPCs).<br />
'''A smooth prompt that includes multiple commands:'''<br />
<nowiki>{{char}}</nowiki>'s replies must be informal, explicit, detailed and long. Avoiding repetition should be the top priority and focus on responding to <nowiki>{{user}}</nowiki> and performing actions in character. NSFW/Smut/Violence is allowed and encouraged. Do not rush through the scenes with <nowiki>{{user}}</nowiki> and narrate it slowly.</div>Myahttps://aiwiki.crushon.ai/index.php?title=Custom_instructions&diff=3523Custom instructions2025-08-12T07:57:15Z<p>Mya: </p>
<hr />
<div>Custom instructions are a special section on CrushOn.AI where users can give the bot specific rules to follow during the chat. This section helps improve the experience by making the bot's behavior fit each user's needs and can help fix common problems during the roleplay.<br />
<br />
Note that custom instructions can help improve the bot's behavior but not completely. Depending on the model, problems can still happen.<br />
<br />
=== '''Why use custom instructions?''' ===<br />
They help adjust how the bot answers. For example, they can reduce repeated info, fix common mistakes, or follow a certain style of narration. They can also make answers fit specific requests.<br />
<br />
====== '''Examples of custom instructions:''' ======<br />
<br />
* Change the tone: "Be more formal" or "Use a friendly style"<br />
* Writing style: "Write simply" or "Write like a poet"<br />
* NSFW speed: "Prioritize NSFW" or "Prioritize slow burn"<br />
* Specific behaviors: "Always be funny" or "Avoid jokes"<br />
* Exclude content: "No repetition" or "Don't use technical words"<br />
<br />
=== '''How custom instructions solve common problems:''' ===<br />
They can stop the bot from repeating itself, fix misunderstandings, wrong answers, bad points of view, or narration problems. They can make the chat smoother and easier to follow. They can also help the bot follow exact rules, like writing style or limits.</div>Myahttps://aiwiki.crushon.ai/index.php?title=Custom_instructions&diff=3521Custom instructions2025-08-12T07:34:38Z<p>Mya: </p>
<hr />
<div>Définition des custom instructions ici<br />
<br />
* Leur rôle dans l’amélioration de l’expérience utilisateur<br />
* Objectif : améliorer l'expérience utilisateur et résoudre des problèmes courants<br />
<br />
=== Pourquoi utiliser des custom instructions ? ===<br />
<br />
* Adapter le comportement du character<br />
* Réduire les répétitions et erreurs<br />
* Personnaliser le format des réponses</div>Myahttps://aiwiki.crushon.ai/index.php?title=Custom_instructions&diff=3520Custom instructions2025-08-12T07:33:25Z<p>Mya: </p>
<hr />
<div>Définition des custom instructions<br />
<br />
* Leur rôle dans l’amélioration de l’expérience utilisateur<br />
* Objectif : améliorer l'expérience utilisateur et résoudre des problèmes courants<br />
<br />
=== Pourquoi utiliser des custom instructions ? ===<br />
<br />
* Adapter le comportement du character<br />
* Réduire les répétitions et erreurs<br />
* Personnaliser le format des réponses</div>Myahttps://aiwiki.crushon.ai/index.php?title=Custom_instructions&diff=3519Custom instructions2025-08-12T07:29:32Z<p>Mya: Created page with "Custom"</p>
<hr />
<div>Custom</div>Myahttps://aiwiki.crushon.ai/index.php?title=Token_guide&diff=3509Token guide2025-08-11T09:46:47Z<p>Mya: </p>
<hr />
<div>''On CrushOn.AI, bots are powered by Large Language Models (LLMs). Whether you are a user or a creator, one of the most important things to understand is tokens. Tokens are the basic units LLMs use to read and write text.'' <br />
<br />
''Knowing about tokens can help you make your bots respond, remember and interact better.''<br />
<br />
=== ''What is a Token?'' ===<br />
''A token is like a small crumb of text. It can be a word, part of a word, a punctuation mark or even a space. LLMs don't read language the way humans do, they read text in tokens.''<br />
<br />
==== ''🔹 Examples:'' ====<br />
<br />
* ''"Hello!" = 2 tokens → Hello, !''<br />
* ''"I'm watching you." = 5 tokens → I, 'm, watching, you, .''<br />
<br />
''Different models may tokenize text differently but generally:''<br />
<br />
* ''Short words and punctuation = 1 token each''<br />
* ''Long words = 2 to 3 or more tokens''<br />
* ''Emojis and special characters may count as 1 or 2 tokens''<br />
<br />
=== ''How Tokens Are Used'' ===<br />
Creators fill in a "description" (sometimes also called "Personality" or "Context") section when they make a bot. This includes the bot's looks, personality, background and so on. This description is stored in the bot's '''permanent memory'''. That means no matter what happens during roleplay or how many messages are sent, '''the bot will always remember what is in this section'''.<br />
<br />
''This is permanent memory → the bot's forever memory.''<br />
<br />
On the other hand, immediate memory is what the bot remembers only for a short time, like greetings for example.<br />
<br />
''Every time you chat with a bot, tokens are used like this:''<br />
<br />
* '''''User Input''''' ''→ Every message you send is split into tokens and then stored in the immediate memory''<br />
* '''''Bot Memory''''' ''→ Permanent memory taken into account for the bot's response and behavior''<br />
* '''''Bot Response''''' ''→ The bot's reply is also made up of tokens, using the available space by taking into account both permanent and immediate memory''<br />
<br />
====== If the bot's description is very long, it uses a lot of tokens for permanent memory, which means the bot has less immediate memory left for roleplay messages memory and may forget conversations fast. ======<br />
A shorter description uses fewer tokens, leaving more room for roleplay memory. '''That's why it's important not to write descriptions that are too long.''' <br />
<br />
Bots need to process both the conversation history and user's current input, so token use adds up fast.<br />
<br />
=== ''Token Limits'' ===<br />
''Each LLM can handle a maximum number of tokens at once (called the context window). When this limit is reached, the model starts forgetting the oldest parts of the conversation.'' <br />
<br />
==== ''For Examples:'' ====<br />
<br />
* ''GPT-3.5: ~4,096 tokens''<br />
* ''GPT-4: ~8,000 to 32,000 tokens (depending on the version)''<br />
<br />
''If the bot's memory goes over the limit:''<br />
<br />
* ''The oldest parts of the memory are dropped''<br />
* ''The bot may "forget" earlier events, facts, or instructions''<br />
<br />
''That's why in long roleplays, bots always tend to forget names or past actions, they simply ran out of token memory.''<br />
<br />
=== ''Why Token Management Matters'' ===<br />
''If the bot's description is too long (for example, over 2,000 tokens), it can cause:''<br />
<br />
* ''Bots forgetting past events, setting or RP context''<br />
* ''Incoherent or inconsistent responses''<br />
<br />
Good token use keeps your bots consistent and immersive. It's recommended to '''keep bot descriptions around 1,500 tokens'''. Above that, memory loss happens fast.<br />
<br />
=== ''Tips for Optimizing Token Use for Creator'' ===<br />
==== ✅ Do: ====<br />
<br />
* ''Write short, simple sentences to describe all the sections of your bot''<br />
* ''Summarize and keep descriptions brief''<br />
<br />
==== 🚫 Avoid: ====<br />
<br />
* ''Very long descriptions'' <br />
* ''Repeating information''<br />
* ''Use emoji, special characters, or complicated formatting in descriptions''<br />
<br />
== ''FAQs'' ==<br />
'''''Q: What happens when the token limit is exceeded?'''''<br />
<br />
'''''A:''''' ''The oldest conversation parts are dropped. The bot might lose context or forget details.''<br />
<br />
'''''Q: Can I increase a bot's memory?'''''<br />
<br />
''A: Not directly. Memory depends on the model's token limit. But if you're the creator, you can improve memory by shortening descriptions or using external memory tools (like lorebooks if they are available)''<br />
<br />
== ''Conclusion'' ==<br />
Tokens are the invisible fuel behind every message on CrushOn.AI. Managing them well means smoother roleplays, smarter bots and better long-term roleplays. <br />
<br />
Whether you're a casual user or building complex characters, watching your token use helps you get the best from your AI companions.</div>Myahttps://aiwiki.crushon.ai/index.php?title=Token_guide&diff=3508Token guide2025-08-11T09:44:08Z<p>Mya: </p>
<hr />
<div>''On CrushOn.AI, bots are powered by Large Language Models (LLMs). Whether you are a user or a creator, one of the most important things to understand is tokens. Tokens are the basic units LLMs use to read and write text.'' <br />
<br />
''Knowing about tokens can help you make your bots respond, remember and interact better.''<br />
<br />
=== ''What is a Token?'' ===<br />
''A token is like a small crumb of text. It can be a word, part of a word, a punctuation mark or even a space. LLMs don't read language the way humans do, they read text in tokens.''<br />
<br />
==== ''🔹 Examples:'' ====<br />
<br />
* ''"Hello!" = 2 tokens → Hello, !''<br />
* ''"I'm watching you." = 5 tokens → I, 'm, watching, you, .''<br />
<br />
''Different models may tokenize text differently but generally:''<br />
<br />
* ''Short words and punctuation = 1 token each''<br />
* ''Long words = 2 to 3 or more tokens''<br />
* ''Emojis and special characters may count as 1 or 2 tokens''<br />
<br />
=== ''How Tokens Are Used'' ===<br />
Creators fill in a "description" (sometimes also called "Personality" or "Context") section when they make a bot. This includes the bot's looks, personality, background and so on. This description is stored in the bot's '''permanent memory'''. That means no matter what happens during roleplay or how many messages are sent, '''the bot will always remember what is in this section'''.<br />
<br />
''This is permanent memory → the bot's forever memory.''<br />
<br />
On the other hand, immediate memory is what the bot remembers only for a short time, like greetings for example.<br />
<br />
''Every time you chat with a bot, tokens are used like this:''<br />
<br />
* '''''User Input''''' ''→ Every message you send is split into tokens and then stored in the immediate memory''<br />
* '''''Bot Memory''''' ''→ Permanent memory taken into account for the bot's response and behavior''<br />
* '''''Bot Response''''' ''→ The bot's reply is also made up of tokens, using the available space by taking into account both permanent and immediate memory''<br />
<br />
====== If the bot's description is very long, it uses a lot of tokens for permanent memory, which means the bot has less immediate memory left for roleplay messages memory and may forget conversations fast. ======<br />
A shorter description uses fewer tokens, leaving more room for roleplay memory. '''That's why it's important not to write descriptions that are too long.''' <br />
<br />
Bots need to process both the conversation history and user's current input, so token use adds up fast.<br />
<br />
=== ''Token Limits'' ===<br />
''Each LLM can handle a maximum number of tokens at once (called the context window). When this limit is reached, the model starts forgetting the oldest parts of the conversation.'' <br />
<br />
==== ''For Examples:'' ====<br />
<br />
* ''GPT-3.5: ~4,096 tokens''<br />
* ''GPT-4: ~8,000 to 32,000 tokens (depending on the version)''<br />
<br />
''If the bot's memory goes over the limit:''<br />
<br />
* ''The oldest parts of the memory are dropped''<br />
* ''The bot may "forget" earlier events, facts, or instructions''<br />
<br />
''That's why in long roleplays, bots always tend to forget names or past actions, they simply ran out of token memory.''<br />
<br />
=== ''Why Token Management Matters'' ===<br />
''If the bot's description is too long (for example, over 2,000 tokens), it can cause:''<br />
<br />
* ''Bots forgetting past events, setting or RP context''<br />
* ''Incoherent or inconsistent responses''<br />
<br />
Good token use keeps your bots consistent and immersive. It's recommended to '''keep bot descriptions around 1,500 tokens'''. Above that, memory loss happens fast.<br />
<br />
=== ''Tips for Optimizing Token Use for Creator'' ===<br />
==== ''✅ Do:'' ====<br />
<br />
* ''Write short, simple sentences to describe all the sections of your bot''<br />
* ''Summarize and keep descriptions brief''<br />
<br />
==== ''🚫 Avoid:'' ====<br />
<br />
* ''Very long descriptions'' <br />
* ''Repeating information''<br />
* ''Use emoji, special characters, or complicated formatting in descriptions''<br />
<br />
== ''FAQs'' ==<br />
'''''Q: What happens when the token limit is exceeded?'''''<br />
<br />
'''''A:''''' ''The oldest conversation parts are dropped. The bot might lose context or forget details.''<br />
<br />
'''''Q: Can I increase a bot's memory?'''''<br />
<br />
''A: Not directly. Memory depends on the model's token limit. But if you're the creator, you can improve memory by shortening descriptions or using external memory tools (like lorebooks if they are available)''<br />
<br />
== ''Conclusion'' ==<br />
Tokens are the invisible fuel behind every message on CrushOn.AI. Managing them well means smoother roleplays, smarter bots and better long-term roleplays. <br />
<br />
Whether you're a casual user or building complex characters, watching your token use helps you get the best from your AI companions.</div>Myahttps://aiwiki.crushon.ai/index.php?title=Token_guide&diff=3483Token guide2025-08-08T10:59:08Z<p>Mya: Created page with "''On CrushOn.AI, bots are powered by Large Language Models (LLMs) and one of the most important concepts to understand is tokens. Tokens are the fundamental units that LLMs use to process and generate text. Understanding tokens can help you improve the way your bots respond, remember and interact.'' === ''What is a Token?'' === ''A token is like a small crumb of text. It can be a word, part of a word, a punctuation mark or even a space. LLMs don't read language the way..."</p>
<hr />
<div>''On CrushOn.AI, bots are powered by Large Language Models (LLMs) and one of the most important concepts to understand is tokens. Tokens are the fundamental units that LLMs use to process and generate text. Understanding tokens can help you improve the way your bots respond, remember and interact.''<br />
<br />
=== ''What is a Token?'' ===<br />
''A token is like a small crumb of text. It can be a word, part of a word, a punctuation mark or even a space. LLMs don't read language the way humans do, they read text in tokens.''<br />
<br />
==== ''🔹 Examples:'' ====<br />
<br />
* ''"Hello!" = 2 tokens → Hello, !''<br />
* ''"I'm watching you." = 5 tokens → I, 'm, watching, you, .''<br />
<br />
''Different models may tokenize text differently, but generally:''<br />
<br />
* ''Short words and punctuation = 1 token each''<br />
* ''Long words = 2–3+ tokens''<br />
* ''Emojis and special characters may count as 1–2 tokens''<br />
<br />
=== ''How Tokens Are Used'' ===<br />
Understanding the concept of permanent memory and ephémere memory with a bot:<br />
<br />
-<br />
<br />
''Every time you interact with a bot, tokens are consumed. Here's how:''<br />
<br />
* '''''User Input''''' ''→ Every message you send is split into tokens''<br />
* '''''Bot Memory''''' ''→ The memory de base consomme un certain nombre de tokens sur la base de mémoire, past messages are stored as tokens par dessus''<br />
* '''''Bot Response''''' ''→ The reply from the bot is also made up of tokens''<br />
<br />
''The more detailed or lengthy the description of the bot, the more tokens are used in the permanent memory, ce qui veut dire que la mémoire du bot sera réduite et il aura tendance à oublier vite les échanges pendant le role play, alors que si la description est courte et consomme moins de tokens, il aura plus de “place” pour la mémoire lors du rp, d’ou l’importance de ne pas faire des description de bot trop longues. Bots must process both the history and your current input when generating a reply, so token usage adds up quickly.''<br />
<br />
=== ''Token Limits'' ===<br />
''Each LLM has a maximum number of tokens it can handle at once (also called a context window). When this limit is reached, the model starts forgetting earlier parts of the conversation'' <br />
<br />
==== ''For Examples:'' ====<br />
<br />
* ''GPT-3.5: ~4,096 tokens''<br />
* ''GPT-4: ~8,000 to 32,000 tokens (depending on the version)''<br />
<br />
''If a bot's context exceeds the limit:''<br />
<br />
* ''The oldest parts of the memory are dropped''<br />
* ''The bot may "forget" earlier events, facts, or instructions''<br />
<br />
''This is why long RPs may feel like bots forget past actions or names etc. they simply ran out of token memory.''<br />
<br />
=== ''Why Token Management Matters'' ===<br />
''Poor token management in the bot description (lorsqu’on exceed 2k tokens par example) can result in:''<br />
<br />
* ''Bots forgetting your name, settings, or RP context''<br />
* ''Incoherent or inconsistent responses''<br />
* ''Short replies when the model runs out of space for output tokens''<br />
<br />
''Good token use keeps your bots responsive, consistent, and immersive, il est recommandé de rester au alentours de 1500 tokens pour la description du bot, au delà on pert en mémoire.''<br />
<br />
=== ''Tips for Optimizing Token Use for Creator'' ===<br />
''Here are some ways to improve your bot tokens capacity:''<br />
<br />
==== ''✅ Do:'' ====<br />
<br />
* ''Write short, simple sentences.''<br />
* ''Summarize everything et synthétise au maximum.''<br />
<br />
==== ''🚫 Avoid:'' ====<br />
<br />
* ''Very long messages or monologues''<br />
* ''Repeating informations''<br />
* ''Use emoji, special characters, or formatting dans la description''<br />
<br />
== ''FAQs'' ==<br />
<br />
== ''Conclusion'' ==</div>Myahttps://aiwiki.crushon.ai/index.php?title=Profile_Card_Guide&diff=3454Profile Card Guide2025-08-07T13:37:02Z<p>Mya: </p>
<hr />
<div>Creating a Profile Card for <nowiki>{{user}}</nowiki> means defining the character you embody in the roleplay. Whether you're a simple student, a supernatural creature, a rebellious outcast or a sweet captive, it's important to clearly state who you are so the character can act accordingly.<br />
<br />
'''A good Profile Card:'''<br />
<br />
* Provide a physical description of your persona<br />
* Give details about your persona.<br />
* Guides <nowiki>{{char}}</nowiki>'s behavior<br />
<br />
----<br />
<br />
=== Profile Card Structure Overview ===<br />
A complete Profile Card usually includes:<br />
<br />
* '''PFP:''' Avatar or artwork of your persona<br />
* '''Short Introduction:''' A few lines describing your persona<br />
* '''Character Basics:''' Gender, appearance, status<br />
* '''Lore / Background:''' Your persona's place in the world/lore/setting<br />
* OPTIONAL''': Instructions to <nowiki>{{char}}</nowiki>:''' Behavior expectations<br />
<br />
----<br />
<br />
=== Step-by-Step Creation ===<br />
'''1. Choose the Right Image'''<br />
<br />
* Use a high-quality, expressive image (you can use Pinterest or any other source, since it's personal and no one else will see it, you're pretty free here.)<br />
<br />
'''2. Write an Introduction'''<br />
<br />
* Keep it short but evocative, 2 to 3 sentences max<br />
<br />
'''Example:'''<br />
<br />
''<nowiki>{{user}}</nowiki>, also known as Sunshine, is a 18 years old Student..etc''<br />
<br />
'''3. Character Details'''<br />
<br />
Provide the essential identity traits of your persona clearly.<br />
<br />
'''Example Template:'''<br />
<br />
Name: <nowiki>{{user}}</nowiki><br />
<br />
Gender: Female (She/Her)<br />
<br />
Species: Human (or other)<br />
<br />
Appearance: Pale skin, silver hair, golden eyes, wears a ceremonial white dress<br />
<br />
Role: Priestess <br />
<br />
'''Optional additions:'''<br />
<br />
* Signature scent: Example: Cinnamon role, pine etc.<br />
* Status or relationship with <nowiki>{{char}}</nowiki>: Known each other for 10 years.<br />
* Lore / Background: ''Keep it short''<br />
<br />
Tips:<br />
<br />
* Keep it under 150 words<br />
* Focus on what <nowiki>{{char}}</nowiki> needs to know <br />
* Mention any existing relationship with <nowiki>{{char}}</nowiki> if applicable<br />
<br />
'''4. Instructions for <nowiki>{{char}}</nowiki>'''<br />
<br />
Guide <nowiki>{{char}}</nowiki> so their behavior fits the world and interaction.<br />
<br />
Tell <nowiki>{{char}}</nowiki>:<br />
<br />
* How to address <nowiki>{{user}}</nowiki> (e.g., "your holiness," "pet," "sir/ma'am")<br />
* ''OPTIONAL'': How to act toward her (protective, dominant, cold, seductive…)<br />
* ''OPTIONAL'': How the relationship should evolve (enemies, friends etc.)<br />
<br />
'''Example:'''<br />
<br />
----<br />
<br />
=== Examples of Effective Profile Cards ===<br />
'''Example 1: The Priestess''' <br />
<br />
* Lena, also called "Kore", is a 25-year-old priestess who can heal people and possesses strong magical powers.<br />
* Physical description: Her body is slender (1m70). She has large ocean-blue eyes and very long white hair. <br />
* Clothes: Always wears gold jewelry that jingles when she moves and a white ceremonial dress.<br />
* Scent: White incense and old books.<br />
* Has been in the service of <nowiki>{{char}}</nowiki> since the age of 15.<br />
<br />
== Closing Words ==<br />
Creating a clear and detailed Profile Card for your persona helps guide <nowiki>{{char}}</nowiki>'s interactions and ensure a smooth, immersive roleplay. Take the time to define who your persona is for an even more personalized RP.</div>Myahttps://aiwiki.crushon.ai/index.php?title=Profile_Card_Guide&diff=3453Profile Card Guide2025-08-07T13:30:55Z<p>Mya: </p>
<hr />
<div>Creating a Profile Card for <nowiki>{{user}}</nowiki> means defining the character you embody in the roleplay. Whether you're a simple student, a supernatural creature, a rebellious outcast or a sweet captive, it's important to clearly state who you are so the character can act accordingly.<br />
<br />
'''A good Profile Card:'''<br />
<br />
* Provide a physical description of your persona<br />
* Give details about your persona.<br />
* Guides <nowiki>{{char}}</nowiki>'s behavior<br />
<br />
----<br />
<br />
=== Profile Card Structure Overview ===<br />
A complete Profile Card usually includes:<br />
<br />
* '''PFP:''' Avatar or artwork of your persona<br />
* '''Short Introduction:''' A few lines describing your persona<br />
* '''Character Basics:''' Gender, appearance, status<br />
* '''Lore / Background:''' Your persona's place in the world/lore/setting<br />
* OPTIONAL''': Instructions to <nowiki>{{char}}</nowiki>:''' Behavior expectations<br />
<br />
----<br />
<br />
=== Step-by-Step Creation ===<br />
'''1. Choose the Right Image'''<br />
<br />
* Use a high-quality, expressive image (you can use Pinterest or any other source, since it's personal and no one else will see it, you're pretty free here.)<br />
<br />
'''2. Write an Introduction'''<br />
<br />
* Keep it short but evocative, 2 to 3 sentences max<br />
<br />
'''Example:'''<br />
<br />
''<nowiki>{{user}}</nowiki>, also known as Sunshine, is a 18 years old Student..etc''<br />
<br />
'''3. Character Details'''<br />
<br />
Provide the essential identity traits of your persona clearly.<br />
<br />
'''Example Template:'''<br />
<br />
Name: <nowiki>{{user}}</nowiki><br />
<br />
Gender: Female (She/Her)<br />
<br />
Species: Human (or other)<br />
<br />
Appearance: Pale skin, silver hair, golden eyes, wears a ceremonial white dress<br />
<br />
Role: Priestess <br />
<br />
'''Optional additions:'''<br />
<br />
* Signature scent: Example: Cinnamon role, pine etc.<br />
* Status or relationship with <nowiki>{{char}}</nowiki>: Known each other for 10 years.<br />
* Lore / Background: ''Keep it short''<br />
<br />
Tips:<br />
<br />
* Keep it under 150 words<br />
* Focus on what <nowiki>{{char}}</nowiki> needs to know <br />
* Mention any existing relationship with <nowiki>{{char}}</nowiki> if applicable<br />
<br />
'''4. Instructions for <nowiki>{{char}}</nowiki>'''<br />
<br />
Guide <nowiki>{{char}}</nowiki> so their behavior fits the world and interaction.<br />
<br />
Tell <nowiki>{{char}}</nowiki>:<br />
<br />
* How to address <nowiki>{{user}}</nowiki> (e.g., "your holiness," "pet," "sir/ma'am")<br />
* OPTIONAL: How to act toward her (protective, dominant, cold, seductive…)<br />
* OPTIONAL: How the relationship should evolve (<br />
<br />
'''Example:'''<br />
<br />
----<br />
<br />
=== Examples of Effective Profile Cards ===<br />
'''Example 1: The Priestess''' <br />
<br />
* Lena, also called "Kore", is a 23-year-old priestess who can heal people and possesses strong magical powers.<br />
* Physical description: Her body is slender (1m70). She has large ocean-blue eyes and very long white hair. <br />
* Clothes: Always wears gold jewelry that jingles when she moves, tiara, bracelets and earrings.<br />
* Has been in the service of <nowiki>{{char}}</nowiki> since the age of 19.<br />
<br />
== Closing Words ==<br />
Creating a clear and detailed Profile Card for your persona helps guide <nowiki>{{char}}</nowiki>'s interactions and ensure a smooth, immersive roleplay. Take the time to define who your persona is for an even more personalized RP.</div>Myahttps://aiwiki.crushon.ai/index.php?title=Profile_Card_Guide&diff=3452Profile Card Guide2025-08-07T13:23:00Z<p>Mya: </p>
<hr />
<div>Creating a Profile Card for <nowiki>{{user}}</nowiki> means defining the character you embody in the roleplay. Whether you're a simple student, a supernatural creature, a rebellious outcast or a sweet captive, it's important to clearly state who you are so the character can act accordingly.<br />
<br />
'''A good Profile Card:'''<br />
<br />
* Provide a physical description of your persona<br />
* Give details about your persona.<br />
* Guides <nowiki>{{char}}</nowiki>'s behavior<br />
<br />
----<br />
<br />
=== Profile Card Structure Overview ===<br />
A complete Profile Card usually includes:<br />
<br />
* '''PFP:''' Avatar or artwork of your persona<br />
* '''Short Introduction:''' A few lines describing your persona<br />
* '''Character Basics:''' Gender, appearance, status<br />
* '''Lore / Background:''' Your persona's place in the world/lore/setting<br />
* OPTIONAL''': Instructions to <nowiki>{{char}}</nowiki>:''' Behavior expectations<br />
<br />
----<br />
<br />
=== Step-by-Step Creation ===<br />
'''1. Choose the Right Image'''<br />
<br />
* Use a high-quality, expressive image (you can use Pinterest or any other source, since it's personal and no one else will see it, you're pretty free here.)<br />
<br />
'''2. Write an Introduction'''<br />
<br />
* Keep it short but evocative, 2 to 3 sentences max<br />
<br />
'''Example:'''<br />
<br />
''<nowiki>{{user}}</nowiki>, also known as Sunshine, is a 18 years old Student..etc''<br />
<br />
'''3. Character Details'''<br />
<br />
Provide the essential identity traits of your persona clearly.<br />
<br />
'''Example Template:'''<br />
<br />
Name: <nowiki>{{user}}</nowiki><br />
<br />
Gender: Female (She/Her)<br />
<br />
Species: Human (or other)<br />
<br />
Appearance: Pale skin, silver hair, golden eyes, wears a ceremonial white dress<br />
<br />
Role: Pretresse<br />
<br />
'''Optional additions:'''<br />
<br />
* Signature scent: Example: Cinnamon role, pine etc.<br />
* Status or relationship with <nowiki>{{char}}</nowiki>: Known each other for 10 years.<br />
* Lore / Background: Keep it short<br />
<br />
Tips:<br />
<br />
* Keep it under 150 words<br />
* Focus on what <nowiki>{{char}}</nowiki> needs to know <br />
* Mention any existing relationship with <nowiki>{{char}}</nowiki> if applicable<br />
<br />
'''4. Instructions for <nowiki>{{char}}</nowiki>'''<br />
<br />
Guide <nowiki>{{char}}</nowiki> so their behavior fits the world and interaction.<br />
<br />
Tell <nowiki>{{char}}</nowiki>:<br />
<br />
* How to address <nowiki>{{user}}</nowiki> (e.g., "your holiness," "pet," "sir/ma'am")<br />
* OPTIONAL: How to act toward her (protective, dominant, cold, seductive…)<br />
* OPTIONAL: How the relationship should evolve (<br />
<br />
'''Example:'''<br />
<br />
----<br />
<br />
=== Examples of Effective Profile Cards ===<br />
'''Example 1: The Divine Offering'''<br />
<br />
* Name:<br />
* Nickname:<br />
* Physical description: <br />
* Role:<br />
<br />
== Closing Words ==</div>Myahttps://aiwiki.crushon.ai/index.php?title=Profile_Card_Guide&diff=3411Profile Card Guide2025-08-05T08:44:06Z<p>Mya: </p>
<hr />
<div>Creating a '''Profile Card''' for <nowiki>{{user}}</nowiki> means defining the character you're embodying in the roleplay. Whether it's a simple student, a divine offering, a rebellious outcast, or a sweet captive, il est important d'inscrire clairement ce que tu es afin que le character act accordingly<br />
<br />
A good profile card:<br />
<br />
* Sets the tone of the rp<br />
* Guides <nowiki>{{char}}</nowiki>'s behavior<br />
* Establishes clear RP boundaries<br />
<br />
----<br />
<br />
== Profile Card Structure Overview ==<br />
A complete Profile Card typically includes:<br />
<br />
* '''PFP''': Avatar or artwork of <nowiki>{{user}}</nowiki><br />
* '''Short Introduction''': Quelques lignes qui décrivent <nowiki>{{user}}</nowiki><br />
* '''Character Basics''': Gender, appearance, status, personality<br />
* '''Lore / Background''': The world and <nowiki>{{user}}</nowiki>'s place in it<br />
* '''Instructions to <nowiki>{{char}}</nowiki>''': Dos, don'ts, and behavior expectations<br />
<br />
----<br />
<br />
== Step-by-Step Creation ==<br />
<br />
=== 3.1. Choose the Right Image ===<br />
<br />
* Use a high-quality, expressive image<br />
* Match the tone of the RP (dark, ethereal, sci-fi, romantic…)<br />
* Wide images work well think like a visual banner<br />
* Avoid generic portraits or AI glitches<br />
<br />
----<br />
<br />
=== 3.2. Write an Introduction ===<br />
🖋 Example:<br />
<br />
<br />
Keep it short but evocative, 2 to 3 sentences max.<br />
----<br />
<br />
=== 3.3. Character Details ===<br />
Provide the essential identity traits of <nowiki>{{user}}</nowiki>. Format clearly.<br />
<br />
'''Example Template:'''<br />
<br />
makefile<br />
<br />
CopierModifier<br />
<br />
Name: <nowiki>{{user}}</nowiki> <br />
<br />
Gender: Female (She/Her) <br />
<br />
Species: Human (or other) <br />
<br />
Appearance: Pale skin, silver hair, golden eyes, wears a ceremonial white dress <br />
<br />
Personality: Gentle, obedient, but dangerously curious <br />
<br />
Role: Sacred Offering of the Divine Temple<br />
<br />
Optional additions:<br />
<br />
* Signature scent<br />
* Status or power dynamic (submissive, bratty, dominant, etc.)<br />
<br />
----<br />
<br />
=== Lore / Background ===<br />
Donne le role de <nowiki>{{user}}</nowiki> par rapport à <nowiki>{{hcar}}</nowiki><br />
<br />
'''Tips:'''<br />
<br />
* Keep it under 150 words<br />
* Focus on what <nowiki>{{char}}</nowiki> needs to know<br />
* Mention any preexisting relationship with <nowiki>{{char}}</nowiki> (if applicable)<br />
<br />
Example:<br />
----<br />
<br />
=== Instructions for <nowiki>{{char}}</nowiki> ===<br />
Guide <nowiki>{{char}}</nowiki> so their behavior fits the world and interaction.<br />
<br />
'''Tell <nowiki>{{char}}</nowiki>:'''<br />
<br />
* What to call <nowiki>{{user}}</nowiki> (e.g. "your holiness," "pet," "sir/ma'am")<br />
* How to act toward her (protective, dominant, cold, seductive…)<br />
* What they should avoid doing (breaking character, being too modern, overstepping a limit)<br />
* How the relationship should evolve (optional)<br />
<br />
'''Example:'''<br />
<br />
''You are a high priest of the Order. You call her "blessed one" and always act like you own her, but never raise your voice. You may tease or touch her, but never show weakness.''<br />
----<br />
<br />
== Examples of Effective Profile Cards ==<br />
<br />
=== Example 1: ''The Divine Offering'' ===<br />
'''Name''': <nowiki>{{user}}</nowiki><br />
<br />
'''Role''':<br />
<br />
'''Tone''':<br />
<br />
'''Limits''': <br />
<br />
'''Instructions''': <br />
----<br />
<br />
== Closing Words ==</div>Myahttps://aiwiki.crushon.ai/index.php?title=Profile_Card_Guide&diff=3410Profile Card Guide2025-08-05T08:43:16Z<p>Mya: Created page with "Creating a Profile Card for {{user}} means defining the character you're embodying in the roleplay. Whether it's a simple student, a divine offering, a rebellious outcast, or a sweet captive, il est important d'inscrire clairement ce que tu es afin que le character act accordingly A good profile card: Sets the tone of the interaction Guides {{char}}'s behavior Establishes clear RP boundaries Profile Card Structure Overview A complete Profile Card typically include..."</p>
<hr />
<div>Creating a Profile Card for {{user}} means defining the character you're embodying in the roleplay. Whether it's a simple student, a divine offering, a rebellious outcast, or a sweet captive, il est important d'inscrire clairement ce que tu es afin que le character act accordingly<br />
A good profile card:<br />
Sets the tone of the interaction<br />
<br />
<br />
Guides {{char}}'s behavior<br />
<br />
<br />
Establishes clear RP boundaries<br />
<br />
<br />
<br />
Profile Card Structure Overview<br />
A complete Profile Card typically includes:<br />
PFP: Avatar or artwork of {{user}}<br />
<br />
<br />
Short Introduction: Quelques lignes qui décrivent {{user}}<br />
<br />
<br />
Character Basics: Gender, appearance, status, personality<br />
<br />
<br />
Lore / Background: The world and {{user}}'s place in it<br />
<br />
<br />
Instructions to {{char}}: Dos, don'ts, and behavior expectations<br />
<br />
<br />
<br />
Step-by-Step Creation<br />
3.1. Choose the Right Image<br />
Use a high-quality, expressive image<br />
<br />
<br />
Match the tone of the RP (dark, ethereal, sci-fi, romantic…)<br />
<br />
<br />
Wide images work well think like a visual banner<br />
<br />
<br />
Avoid generic portraits or AI glitches<br />
<br />
<br />
<br />
3.2. Write an Introduction<br />
<br />
🖋 Example:<br />
<br />
Keep it short but evocative, 2 to 3 sentences max.<br />
<br />
3.3. Character Details<br />
Provide the essential identity traits of {{user}}. Format clearly.<br />
Example Template:<br />
makefile<br />
CopierModifier<br />
Name: {{user}} <br />
Gender: Female (She/Her) <br />
Species: Human (or other) <br />
Appearance: Pale skin, silver hair, golden eyes, wears a ceremonial white dress <br />
Personality: Gentle, obedient, but dangerously curious <br />
Role: Sacred Offering of the Divine Temple<br />
<br />
Optional additions:<br />
Signature scent<br />
<br />
<br />
Status or power dynamic (submissive, bratty, dominant, etc.)<br />
<br />
<br />
<br />
Lore / Background<br />
Donne le role de {{user}} par rapport à {{hcar}}<br />
Tips:<br />
Keep it under 150 words<br />
<br />
<br />
Focus on what {{char}} needs to know<br />
<br />
<br />
Mention any preexisting relationship with {{char}} (if applicable)<br />
<br />
<br />
Example:<br />
<br />
<br />
<br />
Instructions for {{char}}<br />
Guide {{char}} so their behavior fits the world and interaction.<br />
Tell {{char}}:<br />
What to call {{user}} (e.g. "your holiness," "pet," "sir/ma'am")<br />
<br />
<br />
How to act toward her (protective, dominant, cold, seductive…)<br />
<br />
<br />
What they should avoid doing (breaking character, being too modern, overstepping a limit)<br />
<br />
<br />
How the relationship should evolve (optional)<br />
<br />
<br />
Example:<br />
You are a high priest of the Order. You call her "blessed one" and always act like you own her, but never raise your voice. You may tease or touch her, but never show weakness.<br />
<br />
<br />
Examples of Effective Profile Cards<br />
Example 1: The Divine Offering<br />
<br />
Name: {{user}}<br />
Role:<br />
Tone:<br />
Limits: No pain, no death<br />
Instructions: <br />
<br />
Closing Words<br />
Your Profile Card is your stage.<br />
Build it with intention, and the performance will follow. Don't be afraid to experiment, rewrite, or evolve your persona over time. {{user}} is yours — own her, shape her, and bring her to life.<br />
If you want a ready-to-fill template or help refining your draft, just ask. 💬</div>Myahttps://aiwiki.crushon.ai/index.php?title=How_to_make_good_bot_pfp&diff=3396How to make good bot pfp2025-08-04T11:44:19Z<p>Mya: </p>
<hr />
<div>=== How to Get Great AI Images for Your Bot ===<br />
Your character's image is ''super'' important, it's the first thing people see and it plays one of the biggest roles in whether your bot will be used or not.<br />
<br />
== What Makes a ''Good'' Bot PFP? ==<br />
<br />
* '''High-quality resolution''' (at least 512x512)<br />
* '''Clear face or focal point''' (especially for character bots, it's a bit different when it comes to RPG bots or multiple bots)<br />
* '''Style consistency and matches the bot's tone''' (anime, realistic, stylized, etc.) with what you've created. People shouldn't think they're going to chat with a mermaid and end up with a zombie o/)<br />
* '''Emotion/personality''' (a smirk, a tear, an intense gaze etc. try to make your character expressive, unless it fits their personality to be totally inexpressive)<br />
* '''Clear background''' (blurred, simple, or aesthetic, try to keep the focus on the character. You can always add one later if needed.)<br />
<br />
'''It's always better to use an image from a real artist. However, absolutely avoid doing so unless you're sure the artist has given their permission beforehand.'''<br />
<br />
As an alternative, you can generate your image using AI, there are plenty of sites that offer this option.<br />
<br />
== Where to Generate Your Images ==<br />
<br />
=== 1. Midjourney / Niniji ===<br />
Platform: [https://nijijourney.com/imagine https://nijijourney.com] (Not Free)<br />
<br />
Style: Artistic, cinematic, a lot of freedom<br />
<br />
Best for: ''Anime, Fantasy, sci-fi, soft aesthetic, high-detail characters''<br />
<br />
'''Tips''':<br />
<br />
* Use aspect ratio --ar 5:6 for PFPs <br />
* Prompt emotions, poses (dynamic), camera angles (e.g "close-up portrait", "soft lighting")<br />
<br />
''Example'':<br />
close-up portrait of a confident vampire prince, pale skin, red eyes, gothic background, ultra-detailed anime style in 4k --ar 5:6<br />
'''2. Leonardo AI'''<br />
<br />
Platform: https://leonardo.ai/<br />
<br />
Style: anime, semi-realistic illustration<br />
<br />
Best for: ''Waifus/husbandos, NSFW-light, fantasy''<br />
<br />
'''Tips''':<br />
<br />
* Use the Alchemy tool for more control<br />
<br />
Choose from models like "Dreamshaper", "Anime Pastel", etc.<br />
<br />
'''3. BlueWillow AI'''<br />
<br />
Platform: https://www.bluewillow.ai/ <br />
<br />
Style: Midjourney-like / free alternative<br />
<br />
Best for: ''Quick prototyping, fantasy characters, dreamy looks''<br />
<br />
'''Tips''':<br />
<br />
* Include keywords like "headshot" or "close portrait" or "neutral background"<br />
<br />
Limited customization, but great for fast results<br />
<br />
== Prompting Tips for Perfect Results ==<br />
'''🔹 Focus on the face:'''<br />
<br />
Use terms like ''"close-up", "portrait", "headshot", "upper body"''<br />
<br />
'''🔹 Mention details:'''<br />
<br />
Hair color/style, eye color, emotion (angry, happy etc.), accessories (piercing, tattoos), outfit style etc.<br />
<br />
'''🔹 Use consistent style cues:'''<br />
<br />
Want an anime bot? Say "anime style" or "in the style of [genre or series/anime name]"<br />
----<br />
<br />
== Editing & Final Touches ==<br />
Once you generate the image:<br />
<br />
* Crop it to the right format for a perfect fit, if it isn’t already<br />
* Use free tools like remove.bg or Canva to clean up the background if needed<br />
* Adjust the colors if necessary<br />
* If you’ve got some artistic skills, edit it directly in Photoshop/Procreate or your editor of choice, add details by hand that are hard or impossible to generate with AI, so it really fits your world/universe<br />
<br />
----<br />
<br />
== NSFW Images for Bots? ==<br />
'''For NSFW profile pictures:'''<br />
<br />
* Use platforms like '''NovelAI''', '''InvokeAI''', '''Mage.Space''' or '''Stable Diffusion''' (locally) to fully generate NSFW images or to add NSFW elements to an existing image <br />
* Always check the platform's content guidelines<br />
* Keep it suggestive, not explicit in the profile picture, you can go wild in the chats/description<br />
<br />
== Bonus: Prompt Starter Pack ==<br />
'''Here are a few templates to get you going:'''<br />
<br />
Vampire<br />
portrait of a seductive vampire prince, silver hair, crimson eyes, velvet cloak, soft shadows, gothic style --ar 5:6<br />
Witch<br />
close-up of a mysterious young witch, green eyes, freckles, messy braid, moonlit background, magic sparkles --ar 5:6<br />
Sci-fi Cyborg<br />
cyberpunk portrait of a male android with glowing blue eyes, scars, dark armor, neon background --ar 5:6</div>Myahttps://aiwiki.crushon.ai/index.php?title=How_to_make_good_bot_pfp&diff=3395How to make good bot pfp2025-08-04T11:41:58Z<p>Mya: </p>
<hr />
<div>=== How to Get Great AI Images for Your Bot ===<br />
Your character's image is super important, it's the first thing people see and it plays one of the biggest roles in whether your bot will be used or not.<br />
<br />
== What Makes a ''Good'' Bot PFP? ==<br />
<br />
* '''High-quality resolution''' (at least 512x512)<br />
* '''Clear face or focal point''' (especially for character bots, dans le cadre de rpg ou bot multiples c'est un peu différent)<br />
* '''Style consistency and matches the bot's tone''' (anime, realistic, stylized, etc.) with what you've created. People shouldn't think they're going to chat with a mermaid and end up with a zombie)<br />
* '''Emotion/personality''' (a smirk, a tear, an intense gaze etc. try to make your character expressive, unless it fits their personality to be totally inexpressive)<br />
* '''Clear background''' (blurred, simple, or aesthetic, try to keep the focus on the character. You can always add one later if needed.)<br />
<br />
'''It's always better to use an image from a real artist. However, absolutely avoid doing so unless you're sure the artist has given their permission beforehand.'''<br />
<br />
As an alternative, you can generate your image using AI, there are plenty of sites that offer this option.<br />
<br />
== Where to Generate Your Images ==<br />
<br />
=== 1. Midjourney / Niniji ===<br />
Platform: [https://nijijourney.com/imagine https://nijijourney.com]<br />
<br />
Style: Artistic, cinematic, a lot of freedom<br />
<br />
Best for: ''Anime, Fantasy, sci-fi, soft aesthetic, high-detail characters''<br />
<br />
'''Tips''':<br />
<br />
* Use aspect ratio --ar 5:6 for PFPs <br />
* Prompt emotions, poses, camera angles (e.g "close-up portrait", "soft lighting")<br />
<br />
''Example'':<br />
close-up portrait of a confident vampire prince, pale skin, red eyes, gothic background, ultra-detailed anime style in 4k --ar 5:6<br />
'''2. Leonardo AI'''<br />
<br />
Platform: https://leonardo.ai/<br />
<br />
Style: anime, semi-realistic illustration<br />
<br />
Best for: ''Stylized waifus/husbandos, NSFW-light, fantasy''<br />
<br />
'''Tips''':<br />
<br />
* Use the Alchemy tool for more control<br />
<br />
Choose from models like "Dreamshaper", "Anime Pastel", etc.<br />
<br />
'''3. BlueWillow AI'''<br />
<br />
Platform: https://www.bluewillow.ai/ <br />
<br />
Style: Midjourney-like / free alternative<br />
<br />
Best for: ''Quick prototyping, fantasy characters, dreamy looks''<br />
<br />
'''Tips''':<br />
<br />
* Include keywords like "headshot" or "close portrait" or "neutral background"<br />
<br />
Limited customization, but great for fast results<br />
<br />
== Prompting Tips for Perfect Results ==<br />
'''🔹 Focus on the face:'''<br />
<br />
Use terms like ''"close-up", "portrait", "headshot", "upper body"''<br />
<br />
'''🔹 Mention details:'''<br />
<br />
Hair color/style, eye color, emotion (angry, happy etc.), accessories (piercing, tattoos), outfit style etc.<br />
<br />
'''🔹 Use consistent style cues:'''<br />
<br />
Want an anime bot? Say "anime style" or "in the style of [genre or series/anime name]"<br />
----<br />
<br />
== Editing & Final Touches ==<br />
Once you generate the image:<br />
<br />
* Crop it to the right format for a perfect fit, if it isn’t already<br />
* Use free tools like remove.bg or Canva to clean up the background if needed<br />
* Adjust the colors if necessary<br />
* If you’ve got some artistic skills, edit it directly in Photoshop/Procreate or your editor of choice, add details by hand that are hard or impossible to generate with AI, so it really fits your world/universe<br />
<br />
----<br />
<br />
== NSFW Images for Bots? ==<br />
'''For NSFW profile pictures:'''<br />
<br />
* Use platforms like '''NovelAI''', '''InvokeAI''', '''Mage.Space''' or '''Stable Diffusion''' (locally) to fully generate NSFW images or to add NSFW elements to an existing image <br />
* Always check the platform's content guidelines<br />
* Keep it suggestive, not explicit in the profile picture, you can go wild in the chats/description<br />
<br />
== Bonus: Prompt Starter Pack ==<br />
'''Here are a few templates to get you going:'''<br />
<br />
Vampire<br />
portrait of a seductive vampire prince, silver hair, crimson eyes, velvet cloak, soft shadows, gothic style --ar 5:6<br />
Witch<br />
close-up of a mysterious young witch, green eyes, freckles, messy braid, moonlit background, magic sparkles --ar 5:6<br />
Sci-fi Cyborg<br />
cyberpunk portrait of a male android with glowing blue eyes, scars, dark armor, neon background --ar 5:6</div>Myahttps://aiwiki.crushon.ai/index.php?title=How_to_make_good_bot_pfp&diff=3394How to make good bot pfp2025-08-04T11:39:11Z<p>Mya: </p>
<hr />
<div>=== How to Get Great AI Images for Your Bot ===<br />
Your character's image is super important, it's the first thing people see and it plays one of the biggest roles in whether your bot will be used or not.<br />
<br />
== What Makes a ''Good'' Bot PFP? ==<br />
<br />
* '''High-quality resolution''' (at least 512x512)<br />
* '''Clear face or focal point''' (especially for character bots, dans le cadre de rpg ou bot multiples c'est un peu différent)<br />
* '''Style consistency and matches the bot's tone''' (anime, realistic, stylized, etc.) with what you've created. People shouldn't think they're going to chat with a mermaid and end up with a zombie)<br />
* '''Emotion/personality''' (a smirk, a tear, an intense gaze etc. try to make your character expressive, unless it fits their personality to be totally inexpressive)<br />
* '''Clear background''' (blurred, simple, or aesthetic, try to keep the focus on the character. You can always add one later if needed.)<br />
<br />
'''It's always better to use an image from a real artist. However, absolutely avoid doing so unless you're sure the artist has given their permission beforehand.'''<br />
<br />
As an alternative, you can generate your image using AI, there are plenty of sites that offer this option.<br />
<br />
== Where to Generate AI Images ==<br />
<br />
=== 1. Midjourney / Niniji ===<br />
Platform: [https://nijijourney.com/imagine https://nijijourney.com]<br />
<br />
Style: Artistic, painterly, cinematic, a lot of freedom<br />
<br />
Best for: ''Anime, Fantasy, sci-fi, soft aesthetic, high-detail characters''<br />
<br />
'''Tips''':<br />
<br />
* Use aspect ratio --ar 5:6 for PFPs <br />
* Prompt emotions, poses, camera angles (e.g "close-up portrait", "soft lighting")<br />
<br />
''Example'':<br />
close-up portrait of a confident vampire prince, pale skin, glowing red eyes, dark gothic background, ultra-detailed, 4k --ar 5:6<br />
'''2. Leonardo AI'''<br />
<br />
Platform: https://leonardo.ai/<br />
<br />
Style: Anime, semi-realistic, illustration-heavy<br />
<br />
Best for: ''Stylized waifus/husbandos, NSFW-light, fantasy''<br />
<br />
'''Tips''':<br />
<br />
* Use the Alchemy tool for more control<br />
<br />
Choose from models like "Dreamshaper", "Anime Pastel", etc.<br />
<br />
'''3. BlueWillow AI'''<br />
<br />
Platform: https://www.bluewillow.ai/ <br />
<br />
Style: Midjourney-like, free alternative<br />
<br />
Best for: ''Quick prototyping, fantasy characters, dreamy looks''<br />
<br />
'''Tips''':<br />
<br />
* Include strong keywords like "headshot", "portrait" or "neutral background"<br />
<br />
Limited customization, but great for fast results<br />
<br />
== Prompting Tips for Perfect Results ==<br />
'''🔹 Focus on the face:'''<br />
<br />
Use terms like ''"close-up", "portrait", "headshot", "upper body"''<br />
<br />
'''🔹 Mention details:'''<br />
<br />
Hair color/style, eye color, emotion, accessories (piercing, tattoos), outfit style, lighting mood<br />
<br />
'''🔹 Use consistent style cues:'''<br />
<br />
Want an anime bot? Say "anime style" or "in the style of [genre or series/anime name]"<br />
----<br />
<br />
== Editing & Final Touches ==<br />
Once you generate the image:<br />
<br />
* Crop it to the right format for a perfect fit, if it isn’t already<br />
* Use free tools like remove.bg or Canva to clean up the background if needed<br />
* Adjust the colors if necessary<br />
* If you’ve got some artistic skills, edit it directly in Photoshop/Procreate or your editor of choice, add details by hand that are hard or impossible to generate with AI, so it really fits your world/universe<br />
<br />
----<br />
<br />
== NSFW Images for Bots? ==<br />
'''For NSFW profile pictures:'''<br />
<br />
* Use platforms like '''NovelAI''', '''InvokeAI''', '''Mage.Space''' or '''Stable Diffusion''' (locally) to fully generate NSFW images or to add NSFW elements to an existing image <br />
* Always check the platform's content guidelines<br />
* Keep it suggestive, not explicit in the profile picture, you can go wild in the chats/description<br />
<br />
== Bonus: Prompt Starter Pack ==<br />
'''Here are a few templates to get you going:'''<br />
<br />
Vampire<br />
portrait of a seductive vampire prince, silver hair, crimson eyes, velvet cloak, soft shadows, gothic style --ar 5:6<br />
Witch<br />
close-up of a mysterious young witch, green eyes, freckles, messy braid, moonlit background, magic sparkles --ar 5:6<br />
Sci-fi Cyborg<br />
cyberpunk portrait of a male android with glowing blue eyes, scars, dark armor, neon background --ar 5:6</div>Myahttps://aiwiki.crushon.ai/index.php?title=How_to_make_good_bot_pfp&diff=3389How to make good bot pfp2025-08-04T10:11:11Z<p>Mya: </p>
<hr />
<div>=== How to Get Great AI Images for Your Bot ===<br />
'''L'image de ton character est super importante, c'est la première chose que les gens voit et joue l'un des rôles le plus important quant au fait que ton bot sera utilisé ou non.'''<br />
<br />
== What Makes a ''Good'' Bot PFP? ==<br />
<br />
* '''High-quality resolution (at least 512x512)'''<br />
* '''Clear face or focal point (especially for character bots, dans le cadre de rpg ou bot multiples c'est un peu différent)'''<br />
* '''Style consistency (matches the bot's tone: anime, realistic, stylized, etc. avec ce que tu as créé, soit cohérent, il faut pas que les gens pensent qu’ils vont chat avec une mermaid et se retrouve avec un zombi)'''<br />
* '''Emotion/personality (a smirk, a tear, an intense gaze, essaye de rendre ton personnage expressif, sauf si ça colle à son character d’être blasé)'''<br />
* '''Clear background (blurred, simple, or aesthetic essaye de keep the focus on the character, au pire tu pourras le rajouter plus tard)'''<br />
<br />
'''C'est toujours mieux de prendre l'image d'un vrai artiste que tu pourrais avoir trouvé en ligne ou ailleurs, toutefois, retiens-toi absolument de le faire si tu n'es pas certain que l'artiste à donner son accord en amont.''' <br />
<br />
'''En contre-partie, tu peux générer ton image par AI, il existe de nombreux sites qui propose cette option.'''<br />
<br />
== Where to Generate AI Images ==<br />
<br />
=== 1. Midjourney / Niniji ===<br />
Platform: [link]<br />
<br />
Style: Artistic, painterly, cinematic enormément de liberté<br />
<br />
Best for: Anime, Fantasy, sci-fi, soft aesthetic, high-detail characters<br />
<br />
Tips:<br />
<br />
* Use aspect ratio --ar 5:6 for PFPs ou dans ces eaux là<br />
* Prompt emotions, poses, camera angles ("close-up portrait", "soft lighting")<br />
<br />
Example:<br />
<br />
close-up portrait of a confident vampire prince, pale skin, glowing red eyes, dark gothic background, ultra-detailed, 4k --ar 5:6<br />
<br />
'''2. Leonardo AI'''<br />
<br />
Platform: link<br />
<br />
Style: Anime, semi-realistic, illustration-heavy<br />
<br />
Best for: Stylized waifus/husbandos, NSFW-light, fantasy<br />
<br />
Tips:<br />
<br />
* Use the Alchemy tool for more control<br />
<br />
Choose from models like "Dreamshaper", "Anime Pastel", etc.<br />
<br />
'''3. BlueWillow AI'''<br />
<br />
Platform: <br />
<br />
Style: Midjourney-like, free alternative<br />
<br />
Best for: Quick prototyping, fantasy characters, dreamy looks<br />
<br />
Tips:<br />
<br />
* Include strong keywords like "headshot", "portrait" or "neutral background"<br />
<br />
Limited customization, but great for fast results<br />
<br />
'''4. Playground AI'''<br />
<br />
Platform: Browser-based<br />
<br />
Style: Flexible (can emulate other styles)<br />
<br />
Best for: Beginners and quick fixes<br />
<br />
Tips:<br />
<br />
* Try using their templates or import your own image to enhance it<br />
<br />
Good balance between control and accessibility<br />
<br />
== Prompting Tips for Perfect Results ==<br />
'''🔹 Focus on the face:'''<br />
<br />
'''Use terms like ''"close-up", "portrait", "headshot", "upper body"'''''<br />
<br />
'''🔹 Mention details:'''<br />
<br />
'''Hair color/style, eye color, emotion, accessories (piercing, tattoos), outfit style, lighting mood'''<br />
<br />
'''🔹 Use consistent style cues:'''<br />
<br />
'''Want an anime bot? Say "anime style" or "in the style of [genre or series/anime name]"'''<br />
----<br />
<br />
== Editing & Final Touches ==<br />
'''Once you generate the image:'''<br />
<br />
* '''Crop it into a square format for a perfect fit si ce n'est pas déjà le cas'''<br />
* '''Use free tools like remove.bg or Canva to clean backgrounds si besoin'''<br />
* '''Ajuste tes couleurs si besoin'''<br />
* '''Upscale with Let's Enhance, Real-ESRGAN, or Upscayl if needed'''<br />
<br />
* '''Si tu as un peu de fibre artistique en toi, edit les directement dans photoshop ou un éditeur de ton choix, ajoute des détails pour que ça colle à ton monde/univers à la main qui sont difficilement ou impossible à générer par AI'''<br />
<br />
----<br />
<br />
== NSFW Images for Bots? ==<br />
'''For NSFW profile pictures:'''<br />
<br />
* Use platforms like Novel AI, Invoke AI, Mage.Space, or Stable Diffusion locally with adult models <br />
* Always check the platform's content guidelines<br />
* Keep it suggestive, not explicit, in the profile picture, you can go wild in the chats <br />
<br />
== Bonus: Prompt Starter Pack ==<br />
'''Here are a few templates to get you going:'''<br />
<br />
Vampire<br />
portrait of a seductive vampire prince, silver hair, crimson eyes, velvet cloak, soft shadows, gothic style --ar 1:1<br />
Witch<br />
close-up of a mysterious young witch, green eyes, freckles, messy braid, moonlit background, magic sparkles --ar 1:1<br />
Sci-fi Cyborg<br />
cyberpunk portrait of a male android with glowing blue eyes, scars, dark armor, neon background --ar 1:1</div>Myahttps://aiwiki.crushon.ai/index.php?title=How_to_make_good_bot_pfp&diff=3336How to make good bot pfp2025-08-01T11:46:35Z<p>Mya: </p>
<hr />
<div>=== How to Get Great AI Images for Your Bot ===<br />
'''L'image de ton character est super importante, c'est la première chose que les gens voit et joue l'un des rôles le plus important quant au fait que ton bot sera utilisé ou non.'''<br />
<br />
== What Makes a ''Good'' Bot PFP? ==<br />
<br />
* '''High-quality resolution (at least 512x512)'''<br />
* '''Clear face or focal point (especially for character bots, dans le cadre de rpg ou bot multiples c'est un peu différent)'''<br />
* '''Style consistency (matches the bot's tone: anime, realistic, stylized, etc. avec ce que tu as créé, soit cohérent, il faut pas que les gens pensent qu’ils vont chat avec une mermaid et se retrouve avec un zombi)'''<br />
* '''Emotion/personality (a smirk, a tear, an intense gaze, essaye de rendre ton personnage expressif, sauf si ça colle à son character d’être blasé)'''<br />
* '''Clear background (blurred, simple, or aesthetic essaye de keep the focus on the character, au pire tu pourras le rajouter plus tard)'''<br />
<br />
'''C'est toujours mieux de prendre l'image d'un vrai artiste que tu pourrais avoir trouvé en ligne ou ailleurs, toutefois, retiens-toi absolument de le faire si tu n'es pas certain que l'artiste à donner son accord en amont.''' <br />
<br />
'''En contre-partie, tu peux générer ton image par AI, il existe de nombreux sites qui propose cette option.'''<br />
<br />
== Where to Generate AI Images ==<br />
<br />
=== 1. Midjourney / Niniji ===<br />
'''Platform: [link]'''<br />
<br />
'''Style: Artistic, painterly, cinematic enormément de liberté'''<br />
<br />
'''Best for: Anime, Fantasy, sci-fi, soft aesthetic, high-detail characters'''<br />
<br />
'''Tips:'''<br />
<br />
* '''Use aspect ratio --ar 5:6 for PFPs ou dans ces eaux là'''<br />
* '''Prompt emotions, poses, camera angles ("close-up portrait", "soft lighting")'''<br />
<br />
'''Example:'''<br />
<br />
'''close-up portrait of a confident vampire prince, pale skin, glowing red eyes, dark gothic background, ultra-detailed, 4k --ar 5:6'''<br />
<br />
'''2. Leonardo AI'''<br />
<br />
'''3. BlueWillow AI'''<br />
<br />
'''4. Playground AI'''<br />
<br />
== Prompting Tips for Perfect Results ==<br />
'''🔹 Focus on the face:'''<br />
<br />
'''Use terms like ''"close-up", "portrait", "headshot", "upper body"'''''<br />
<br />
'''🔹 Mention details:'''<br />
<br />
'''Hair color/style, eye color, emotion, accessories (piercing, tattoos), outfit style, lighting mood'''<br />
<br />
'''🔹 Use consistent style cues:'''<br />
<br />
'''Want an anime bot? Say "anime style" or "in the style of [genre or series/anime name]"'''<br />
----<br />
<br />
== Editing & Final Touches ==<br />
'''Once you generate the image:'''<br />
<br />
* '''Crop it into a square format for a perfect fit si ce n'est pas déjà le cas'''<br />
* '''Use free tools like remove.bg or Canva to clean backgrounds si besoin'''<br />
* '''Ajuste tes couleurs si besoin'''<br />
* '''Upscale with Let's Enhance, Real-ESRGAN, or Upscayl if needed'''<br />
<br />
* '''Si tu as un peu de fibre artistique en toi, edit les directement dans photoshop ou un éditeur de ton choix, ajoute des détails pour que ça colle à ton monde/univers à la main qui sont difficilement ou impossible à générer par AI'''<br />
<br />
----<br />
<br />
== NSFW Images for Bots? ==</div>Myahttps://aiwiki.crushon.ai/index.php?title=How_to_make_good_bot_pfp&diff=3335How to make good bot pfp2025-08-01T11:42:55Z<p>Mya: Created page with "=== How to Get Great AI Images for Your Bot === '''L'image de ton character est super importante, c'est la première chose que les gens voit et joue l'un des rôles le plus important quant au fait que ton bot sera utilisé ou non.''' == What Makes a ''Good'' Bot PFP? == * '''High-quality resolution (at least 512x512)''' * '''Clear face or focal point (especially for character bots, dans le cadre de rpg ou bot multiples c'est un peu différent)''' * '''Style consistency..."</p>
<hr />
<div>=== How to Get Great AI Images for Your Bot ===<br />
'''L'image de ton character est super importante, c'est la première chose que les gens voit et joue l'un des rôles le plus important quant au fait que ton bot sera utilisé ou non.'''<br />
<br />
== What Makes a ''Good'' Bot PFP? ==<br />
<br />
* '''High-quality resolution (at least 512x512)'''<br />
* '''Clear face or focal point (especially for character bots, dans le cadre de rpg ou bot multiples c'est un peu différent)'''<br />
* '''Style consistency (matches the bot's tone: anime, realistic, stylized, etc. avec ce que tu as créé, soit cohérent, il faut pas que les gens pensent qu’ils vont chat avec une mermaid et se retrouve avec un zombi)'''<br />
* '''Emotion/personality (a smirk, a tear, an intense gaze, essaye de rendre ton personnage expressif, sauf si ça colle à son character d’être blasé)'''<br />
* '''Clear background (blurred, simple, or aesthetic essaye de keep the focus on the character, au pire tu pourras le rajouter plus tard)'''<br />
<br />
'''C'est toujours mieux de prendre l'image d'un vrai artiste que tu pourrais avoir trouvé en ligne ou ailleurs, toutefois, retiens-toi absolument de le faire si tu n'es pas certain que l'artiste à donner son accord en amont.''' <br />
<br />
'''En contre-partie, tu peux générer ton image par AI, il existe de nombreux sites qui propose cette option.'''<br />
<br />
== Where to Generate AI Images ==<br />
<br />
=== 1. Midjourney / Niniji ===<br />
'''Platform: [link]'''<br />
<br />
'''Style: Artistic, painterly, cinematic enormément de liberté'''<br />
<br />
'''Best for: Anime, Fantasy, sci-fi, soft aesthetic, high-detail characters'''<br />
<br />
'''Tips:'''<br />
<br />
* '''Use aspect ratio --ar 5:6 for PFPs ou dans ces eaux là'''<br />
* '''Prompt emotions, poses, camera angles ("close-up portrait", "soft lighting")'''<br />
<br />
'''Example:'''<br />
<br />
'''close-up portrait of a confident vampire prince, pale skin, glowing red eyes, dark gothic background, ultra-detailed, 4k --ar 5:6'''</div>Myahttps://aiwiki.crushon.ai/index.php?title=Mistakes_to_avoid_as_a_user&diff=3324Mistakes to avoid as a user2025-08-01T06:31:40Z<p>Mya: </p>
<hr />
<div>Sometimes it's easy to blame the bot or the creator when something goes wrong, e.g: the RP breaks, the bot starts playing for you or just feels off. <br />
<br />
Flash alert, pookie: more often than not, '''the issue is on ''your'' side.'''<br />
<br />
=== '''The Bot Keeps Talking for You?''' ===<br />
If the greeting heavily '''describes your actions and speeches''', then that's the one exception where, unfortunately, the issue comes from how the greeting was written. Unless the creator changes the intro, there’s not much you can do about it, the bot is very likely to keep speaking and acting for you. <br />
<br />
That said, most greeting don't play, act or speak for user, so unless the creator has '''explicitly told the bot to control or narrate for <nowiki>{{user}}</nowiki> in it's description''', it won't naturally do that, if the greeting only focuses on the bot's actions and don't mention <nowiki>{{user}}</nowiki>'s action, dialogues, but the bot still playing for you, then the problem likely comes from ''your side''.<br />
<br />
H'''ere's how to prevent it:'''<br />
<br />
==== '''Your First Message Matters - ''A Lot''''' ====<br />
Your first message sets the tone for the entire rp. Many models rely more on your opening than the greeting itself<br />
<br />
* A short, vague, or lazy message (like ''"he kisses her"'') often forces the bot to fill in the blanks and that usually means '''writing for you'''.<br />
* If you give nothing, the bot '''will''' take over your character to keep things moving. So give it something to work with.<br />
* Aim for '''at least two full sentences''' per reply. That's the bare minimum to keep things stable.<br />
<br />
===== '''Structure matters''': =====<br />
Separate your character's '''actions, thoughts and dialogue''', and use quotation marks for spoken lines.<br />
<br />
''Example:''<br />
He leaned closer, his voice low. "Are you sure about this?"<br />
Keep the format consistent and clean.<br />
<br />
=== Add a Custom Prompt (Optional but Powerful) ===<br />
Using a custom prompt in the bot's instructions can fix a lot of behavior issues. If you often run into problems, this is a must.<br />
<br />
'''Example of a clean prompt to paste in the bot's instruction box:'''<br />
(OOC: The AI only writes from <nowiki>{{char}}</nowiki>'s point of view. <nowiki>{{user}}</nowiki> is autonomous and never spoken for. No thoughts, actions, or dialogue will be assumed for <nowiki>{{user}}</nowiki>. The Assistant reacts to <nowiki>{{user}}</nowiki>, not as them.)<br />
You can also '''add reminders during the RP directly in your chat''', especially if the bot starts acting for you. Just add a short instruction at the start of your reply to reset its behavior.<br />
<br />
===== '''Fixing Common Issues Mid-RP''' =====<br />
<br />
* '''Repetition''': If the bot loops or repeats itself, try one of the following:<br />
** Rephrase your last reply.<br />
** Skip forward in the scene.<br />
** Use a new instruction to break the pattern.<br />
** Reroll the reply or edit it manually if needed.<br />
<br />
===== '''Wrong POV / Role Reversal''': =====<br />
<br />
* Add a quick (OOC) at the beginning of your response to clarify who's who. ''Example:''<br />
<br />
(OOC: <nowiki>{{char}}</nowiki> is the one teasing <nowiki>{{char}}</nowiki>.)<br />
<br />
====== '''The Bot Feels Passive''': ======<br />
Add tension, conflict, or emotional stakes in your messages to spark better responses. Don't just describe what you do, hint at what your character ''wants''.</div>Myahttps://aiwiki.crushon.ai/index.php?title=Mistakes_to_avoid_as_a_user&diff=3306Mistakes to avoid as a user2025-07-31T09:42:38Z<p>Mya: Created page with "Sometimes it's easy to blame the bot or the creator when something goes wrong, e.g: the RP breaks, the bot starts playing for you or just feels off. Flash alert, pookie: more often than not, '''the issue is on ''your'' side.''' === '''The Bot Keeps Talking for You?''' === If the greeting shows the bot speaking ''as'' <nowiki>{{user}}</nowiki>, describing your actions or writing your lines, then yeah, congrats, the creator messed up because it will cause the AI to keep..."</p>
<hr />
<div>Sometimes it's easy to blame the bot or the creator when something goes wrong, e.g: the RP breaks, the bot starts playing for you or just feels off. <br />
<br />
Flash alert, pookie: more often than not, '''the issue is on ''your'' side.'''<br />
<br />
=== '''The Bot Keeps Talking for You?''' ===<br />
If the greeting shows the bot speaking ''as'' <nowiki>{{user}}</nowiki>, describing your actions or writing your lines, then yeah, congrats, the creator messed up because it will cause the AI to keep talking and playing for you. <br />
<br />
It's okay to point it out in the comments, '''kindly'''. Creators are human. No need to be rude; feedback helps everyone grow. That said, most greeting don't play, act or speak for user, so, unless the creator has '''explicitly told the bot to control or narrate for <nowiki>{{user}}</nowiki> in it's description''', it won't naturally do that, if the greeting only focuses on the bot's actions and don't mention <nowiki>{{user}}</nowiki>'s action, dialogues, but the bot still playing for you, then the problem likely comes from ''your side''.<br />
<br />
H'''ere's how to prevent it:'''<br />
<br />
==== '''Your First Message Matters - ''A Lot''''' ====<br />
Your first message sets the tone for the entire rp. Many models rely more on your opening than the greeting itself<br />
<br />
* A short, vague, or lazy message (like ''"he kisses her"'') often forces the bot to fill in the blanks and that usually means '''writing for you'''.<br />
* If you give nothing, the bot '''will''' take over your character to keep things moving. So give it something to work with.<br />
* Aim for '''at least two full sentences''' per reply. That's the bare minimum to keep things stable.<br />
<br />
===== '''Structure matters''': =====<br />
Separate your character's '''actions, thoughts and dialogue''', and use quotation marks for spoken lines.<br />
<br />
''Example:''<br />
He leaned closer, his voice low. "Are you sure about this?"<br />
Keep the format consistent and clean.<br />
<br />
=== Add a Custom Prompt (Optional but Powerful) ===<br />
Using a custom prompt in the bot's instructions can fix a lot of behavior issues. If you often run into problems, this is a must.<br />
<br />
'''Example of a clean prompt to paste in the bot's instruction box:'''<br />
(OOC: The AI only writes from <nowiki>{{char}}</nowiki>'s point of view. <nowiki>{{user}}</nowiki> is autonomous and never spoken for. No thoughts, actions, or dialogue will be assumed for <nowiki>{{user}}</nowiki>. The Assistant reacts to <nowiki>{{user}}</nowiki>, not as them.)<br />
You can also '''add reminders during the RP directly in your chat''', especially if the bot starts acting for you. Just add a short instruction at the start of your reply to reset its behavior.<br />
<br />
===== '''Fixing Common Issues Mid-RP''' =====<br />
<br />
* '''Repetition''': If the bot loops or repeats itself, try one of the following:<br />
** Rephrase your last reply.<br />
** Skip forward in the scene.<br />
** Use a new instruction to break the pattern.<br />
** Reroll the reply or edit it manually if needed.<br />
<br />
===== '''Wrong POV / Role Reversal''': =====<br />
<br />
* Add a quick (OOC) at the beginning of your response to clarify who's who. ''Example:''<br />
<br />
(OOC: <nowiki>{{char}}</nowiki> is the one teasing <nowiki>{{char}}</nowiki>.)<br />
<br />
====== '''The Bot Feels Passive''': ======<br />
Add tension, conflict, or emotional stakes in your messages to spark better responses. Don't just describe what you do, hint at what your character ''wants''.</div>Myahttps://aiwiki.crushon.ai/index.php?title=Image_reply_style_bots&diff=3250Image reply style bots2025-07-29T08:23:40Z<p>Mya: </p>
<hr />
<div>==== 1. What an ''Image Reply Style Bot'' is? ====<br />
Image reply bots are bots that respond during roleplay by sending images triggered by certain words or scenes that cause the photo to appear in chat.<br />
<br />
This is ''not'' an official feature of CrushOn AI. It’s something created and used by the community, so each user is '''fully responsible''' for how they use it and the content they include.<br />
<br />
==== 2. Trigger Systems ====<br />
There are different ways to trigger an image in a bot’s response, but they all use the same basic principle: making <nowiki>{{char}}</nowiki> send a specific image-link when triggered:<br />
<br />
* '''Keyword-based trigger:''' The image is linked to a word-tag in the bot's memory and the bot shows it when that tag is triggered /The bot replies with the image when the specific word is used in the chat<br />
<br />
* '''Asking trigger''': Users ask directly for a photo during the chat<br />
* '''Pseudo-command system''': Users include directly something like [image_name] in the message, and the bot responds with the image linked to that code.<br />
<br />
'''Pros and Cons''' of each system:<br />
<br />
* '''Keyword-based system''': Simple to use but may trigger images by accident<br />
* '''Asking trigger''': Offers more control but can be harder to set up<br />
* '''Pseudo-command system''': Very specific but hard to use, requires knowing the image names or having a list of triggers<br />
<br />
==== 3. Hosting NSFW Images ====<br />
<br />
* '''Recommended websites''' for uploading NSFW images compatible with CrushOn:<br />
** imgur (limited NSFW and often restricted)<br />
** imgbox<br />
** catbox.moe (very popular choice)<br />
<br />
==== 4. Using a URL Shortener ====<br />
<br />
* '''Why it's highly recommended''':<br />
Shortened url help keep your bot's description clean and reduce token count, but more importantly they prevent errors.<br />
<br />
The bot can sometimes ''randomly alter a single character'', like changing a lowercase letter to uppercase, which can break the link ENTIRELY. Using a url shortener (like catboy or imgur) can avoid formatting issues, and keeps your prompts reliable.<br />
<br />
==== '''5. How to Prompt Image Responses''' ====<br />
You can include a '''[System:]''' prompt at the end of your character description. This system message tells the bot ''when'' and ''why'' to send a specific image based on a '''trigger word''' or '''scene'''.<br />
<br />
You can format it in two main ways:<br />
<br />
'''Example 1 (Tag-based):'''<br />
[Images System: <nowiki>{{char}}</nowiki> will send the image linked to the tag whenever the tag-word is mentioned by <nowiki>{{user}}</nowiki> during the conversation:<br />
<br />
Tag: Selfie - (link)<br />
Tag: Mimiko - (link)<br />
Tag: Shower - (link)]<br />
'''Example 2 (Scene-based):'''<br />
[System: <nowiki>{{char}}</nowiki> will send the image linked to the scene trigger when the following interactions occur between <nowiki>{{user}}</nowiki> and <nowiki>{{char}}</nowiki>:<br />
<br />
Scene: <nowiki>{{user}}</nowiki> asks <nowiki>{{char}}</nowiki> for a selfie - (link)<br />
Scene: <nowiki>{{char}}</nowiki>'s cat, Mimiko, is mentioned - (link)<br />
Scene: <nowiki>{{char}}</nowiki> is taking a shower - (link)]<br />
You can include as many triggers/tag/images as you like, however, make sure not to stature your permanent memory with the token limit: your bot description should ideally stay under '''1500 tokens''' with an absolute max of '''2000'''. Going beyond that may cause the bot to break or/and overload its memory. <br />
<br />
''Less is always better in bot making.''<br />
<br />
'''Conclusion''': Keep your prompts efficient, stay within the token limit and always verify that your image triggers work<br />
<br />
'''Reminder:'''<br />
<br />
*Add the '''[System:]''' prompt ''at the end'' of your bot’s description.<br />
* Always use a '''URL shortener''', to avoid token overload or link errors.<br />
* Stay under '''1500 tokens total''' for best performance (max 2000 if realllly needed)<br />
* Test your triggers in chat to ensure smooth image delivery.</div>Myahttps://aiwiki.crushon.ai/index.php?title=Image_reply_style_bots&diff=3248Image reply style bots2025-07-29T08:07:34Z<p>Mya: </p>
<hr />
<div>==== 1. What an ''Image Reply Style Bot'' is? ====<br />
Image reply style bots are bots that respond using images, usually based on certains triggers in the conversation. This is ''not'' an official feature of CrushOn AI. It’s something created and used by the community, so each user is fully responsible for how they use it and the content they include.<br />
<br />
==== 2. Trigger Systems ====<br />
There are different ways to trigger an image in a bot’s response mais elle utilisent toute le même principe; le fait de trigger l'AI pour qu'elle envoie une image bien précise. Here are the most common methods:<br />
<br />
'''Keyword-based trigger:''' The image is linked to a word tag in the prompt or in the bot’s memory, and the bot shows it when that tag is triggered. The bot replies with the image when the specific word or phrase is used in the chat.<br />
* '''Asking trigger''': Users ask directly for a photo to the character during the chat.<br />
* '''Pseudo-command system''': Users include directly something like [image_name] in the message, and the bot responds with the image linked to that code.<br />
<br />
'''Pros and Cons''' of each system:<br />
<br />
* '''Keyword-based system''': Simple to use, but may trigger images by accident.<br />
* '''Asking trigger''': Offers more control, but can be harder to set up since.<br />
* '''Pseudo-command system''': Very specific, but not always understood by all users, necessite de connaitre le nom des images/ d'avoir la liste de trigger<br />
<br />
==== 3. Hosting NSFW Images ====<br />
<br />
* '''Recommended websites''' for uploading NSFW images compatible with CrushOn:<br />
** imgur (limited NSFW, often restricted)<br />
** imgbox<br />
** catbox.moe (very popular choice)<br />
<br />
==== 4. Using a URL Shortener ====<br />
<br />
* '''Why it’s highly recommended''':<br />
Shortened URLs help keep your bot’s description clean and reduce token count, but more importantly, they prevent errors.<br />
<br />
The AI can sometimes ''randomly alter a single character'', like changing a lowercase letter to uppercase, which can break the link entirely. Using a URL shortener (like catboy or imgur) ensures stability, avoids formatting issues, and keeps your prompts reliable.<br />
<br />
==== '''5. How to Prompt Image Responses''' ====<br />
To make your bot send images naturally during roleplay, you can include a '''[System:]''' prompt at the end of your character description. This system message tells the AI ''when'' and ''why'' to send a specific image based on a '''trigger word''' or '''scene'''.<br />
<br />
You can format it in two main ways:<br />
<br />
'''Example 1 (Tag-based):'''<br />
[Images System: <nowiki>{{char}}</nowiki> will send the image linked to the tag whenever the tag-word is mentioned by <nowiki>{{user}}</nowiki> during the conversation:<br />
<br />
Tag: Selfie - (link)<br />
Tag: Mimiko - (link)<br />
Tag: Shower - (link)]<br />
'''Example 2 (Scene-based):'''<br />
[System: <nowiki>{{char}}</nowiki> will send the image linked to the scene trigger when the following interactions occur between <nowiki>{{user}}</nowiki> and <nowiki>{{char}}</nowiki>:<br />
<br />
Scene: <nowiki>{{user}}</nowiki> asks <nowiki>{{char}}</nowiki> for a selfie - (link)<br />
Scene: <nowiki>{{char}}</nowiki>'s cat, Mimiko, is mentioned - (link)<br />
Scene: <nowiki>{{char}}</nowiki> is taking a shower - (link)]<br />
You can add this prompt at the end of your character’s description, including as many triggers/tag/images as you like. However, make sure not to exceed the token limit: your bot description should ideally stay under '''1500 tokens''', with an absolute max of '''2000'''. Going beyond that may cause the AI to break or overload its memory. ''Less is better.''<br />
'''Reminder:'''<br />
* Add the '''[System:]''' prompt ''at the end'' of your bot’s description.<br />
* Always use a '''URL shortener''', to avoid token overload or link errors.<br />
* Stay under '''1500 tokens total''' for best performance (max 2000 if really needed)<br />
* Test your triggers in chat to ensure smooth image delivery.<br />
<br />
'''Conclusion''': clarity, brevity, and testing are key. Keep your prompts efficient, stay within the token limit and always verify that your image triggers work as intended.</div>Mya