His Highness Faithy

Welcome Back!~ ヾ( ˃ᴗ˂ )◞ • *✰ To part 2!Last time we went into basics. This part will be more visual ദ്ദി(˵ •̀ ᴗ - ˵ ) ✧We'll continue were we left off last time with checkpoint recommendations (Thank you for your patience)I'll be linking all models, click on them to reach their page, MIDDLE MOUSE BUTTON Click. so it doesn't close this page ( ◺˰◿ )Table Of ContentCommon TerminologyCreate Tool (Text2Image)Generation basicsPrompts In DepthRecommendationsCreations settingsCheckpointsLoraEmbeddingArtist TagsAdvanced Generation tipsOther Create ToolsExploring TensorAdviceRecommendationsCheckpointsCheckpoints will be tested with these following settingsSampler: dpmpp_2s_ancestral Scheduler: Karras Sampler Steps: 20CFG Scale: 6 Clip Skip: 2 Aspect Ratio: SquarePrompt, and Negative Prompt Every Model will be tested.Checkpoint onlyCheckpoint + Adetailer (face_yolov8m.pt, 20 Steps, 0.5 detection, 0.45 Denoising strength)Checkpoint + Lora + EmbeddingCheckpoint + Lora + Embedding + AdetailerLora + Embedding Used:Keep in mind: Several checkpoints look generic alone, but work really well with Lora. While other checkpoints look great alone, but struggle with Lora causing deformity, and green noise to appear in the image.The Lora I'm using is geared more toward 2D than 3D, Let your creativity flow your own combination of checkpoint and Lora ദ്ദി(˵ •̀ ᴗ - ˵ ) ✧Prompt and Lora used are important for the end result, if you want a 3D, game CG, then mention it in the prompt. Several things you can just word instead of using a Lora to represent it (for example a well known character)Anime 2DHomoSimile XL: My favorite!, My love, my breath, my generation air( ˘ ³˘)♥︎ Words can not do it justice. Works really well with 20 steps, and handles the use of multiple loras at the same time smoothly. I don't even need to use adetailer most of the time, gets updated, Love it~ ( ˘ ³˘)♥︎WAI-NSFW-illustrious-SDXL: Pretty Good, has decent style on it's own, you won't struggle with making NSFW content with it, gets updatedIllustrious Lumine: Very stylized on it's own, doesn't handle multiple loras well. 1-3 Loras would work well.Dreamy Illustrious XL:✨ VelvetXL ✨: Good base Male Checkpoint =w=, (gotta support anyone having a bara model<(˘ ˘ ˘)>)Nova Anime XL: This is my roman empire ♡(˃͈ ˂͈ ). The first Checkpoint I've used, the ruler of my heart, experiment with the older version, I go back to V6 regularly. Gets regularly updated.Nova Unreal XL - Crody:Illustrious-Anime StreamsIllustrious-Anime Creators: Honestly surprised by it, had it in my fav but never used it before, it's pretty good.Alchemix Illustrious:✨JANKU v4.0 - NoobAI EPS + RouWei NSFW Illustrious: Mmmm ( ´͈ ٣ `͈) Those Pecs. Huh? right the model. pretty good next!.Shibui-Illustrious:Aphantasia R - RouWei Based Illustrious: decent, the face seem more into semi-realism, pretty unique.Akiraの电影工坊丨movie scenery: Super cute, good for furry gens, boykisser is awaiting you to gen him, an issue that occurs is a square the appears when using adetailer, you clearly see it in the "C L E AD" with the brown face square. pretty good checkpoint, do not use adetailer on itZUKI cute ILLZUKI cute: Y'all see those man-tiddies right!?? (˵ ¬ᴗ¬˵)Echo IllustriousXL: creates smaller proportions, useful if you are going for a Chibi, or a smurf ദ്ദി(˵ •̀ ᴗ - ˵ ) ✧Late winter Anime style: My go to model for chibis, + it's very easy on the eyes, not in an alluring way but comforting ( •̯́ ₃ •̯̀) (slice of life)Extreme Day White Omen - White and Bright Artistic: Clean simple look, creator has other checkpoints that might fit your style, this is my second favorite checkpoint by them, after late winter.JAM IllustriousNAIXL_MMMMix (version 7): version 8 gives a semi-realism and 3D look to an image. version 8 will be tested in the semi realism section.ZFAn_anime: Very aesthetic, good base, wonderful model kudos for the creator ♡(˃͈ ˂͈ )ShinoKira:Flat Minimalism Line Art Illustrious: Good if your going for that screencap style. Old anime style works well with it.Midnight Sakura Illustrious: It's very different in a good way, I hope the creator continues updating it.Sierunami: Sleek, Clean look, recommended is 28 steps, it still looks good with 20. breathe in DO YOU SEE THOSE TIDDIES ≋(￣﹃￣)≋,🎨llustrious - AnimeMaster FFA By NukeA.I🎨Illustrious - MultiStyle Variation A: Sit down, I have beef with this checkpoint. A love-hate relationship. It has a great base, it gets deformity easily with loras (even when using 25 steps) however results of using it with adetailer are chef kiss, Summary it's a great model, amazing even. However it doesn't react well to loras without adetailer.TQ - IllustriousXL v.1.0: it's decent, creator made a style lora specifically for itPrefect illustrious xl: Very soft Look, Squeezable pecs 10/10, the lighting and everything without lora is very soft, with lora it requires adetailer or deformity will appear (granted i am using 6 loras, sooo it could be that too (๑>؂•̀๑))ToxicEchoXL: Aesthetic , watercolor painting style.Anime 2.5 - 3DYou can achieve 2.5-3D with prompt and Lora too.yomama 2.5D [Illustrious] [Pony]Thick, big-headed guy: just the way I like em Who SAID THAT!? ( ˶°ㅁ°) !!🎨Illustrious - MultiStyle Variation A+R:Alpha Sigma Uxion Illustrious | ASUI: Smooth, Wet, Squee- I mean aesthetically well (ᵕ—ᴗ—)AniCoreXL: Add a 2.5-3D Lora and prompt, like "CG, 2.5d" etc, very very VERY good checkpoint. The image under it, I've added "3D" to the promptNova Anime3D XL - Crody: Pic under it has "3D" added to the prompt.RizmiX Wonder Illustrious Anime XL: Like before you'll have to mention some keywords to make it look more 2.5D, Like realistic, 8k, 8HD =w=Semi-Realism: (You can use Noob, and even 2D checkpoints to make 3D, Semi-realism)GX-Illustrious 型男: My absloute favorite model that isn't Homosimile, I Adore It! do y'all see the pecs! ( ¯ ³¯)♡ , the style is so easy on the eyes, and so scrumptious delicious! If the creator ever falls upon my post, First Heya (˶˃ ᵕ ˂˶)~Second I'll make you not one but TWO trays of brownies~SketchZa JuicyLust:NAIXL_MMMMix V8: Keep In Mind V8 is this style while 7 is anime.Prototypes - Illustrious:cherryG:Juicy Majic 🍸 Illustrious:LunarGrapeMix🍇 (Illustrious):Rizmix LuxNova Illustrious V.1 XL:JunkJourney Illustrious:🎨Illustrious - Anime 2.5D Style: There is no escaping Nuke... it's suppose to be Anime, but it's giving more semi-realism.Zephyr Illustrious: Using Adeatiler creates squares framing the faces, beside that, good model.nero: Decent model, Squares framing the face when using Adeatiler for "version 1.3 cinematic" 1.2 is pretty good.CartoonKiwiMix-XL: Really good, use artist prompt to adjust the style.Illustrious Cartoon: Only for Pro users, sorry (╥﹏╥)Furry:(You don't need to specifically use these models, most checkpoints gen anthro, furries)just state it in the prompt (๑ᵔ⤙ᵔ๑). I don't really gen Furries that often. I'll link the furry related Lora here as well)Indigo Furry Mix XL:Mol_Keun Mix (Furry Mix / CuteCrunch):Lora:Furry style for PonyJinusenpai/Jinu Style [Illustrious]Joaoppereiraus Style [Illustrious]Sukebepanda Style | IllustriousDoppel StyleNoob Checkpoints: It's an anime based model that uses the complete Danbooru and e621 datasets here's an amazing Article created by L_A_X from civitIn summary, it's a well trained model, that requires you to choose what you write for the prompt wisely, using booru tags and E621 (consider it a furry danbooru). I find them to be better at implementing Lora than other checkpointsrizMix Noob Illustrious Pastel: My favorite noob checkpoint, as the name suggests it's uses more muted colours.Madly Mix - Night Noob: Pretty forgiving if you don't know what to write, pretty on it's own~coco-Style-XL: more of a semi-realism, 3D look to it.NTR MIX | illustrious-XL | Noob-XL V4: More stable than XIIINTR MIX | illustrious-XL | Noob-XL XIII: I've seen in special cases, increasing the number of steps to be between 28-35 helps the image clear out from deformity, would not recommend up scaling to 2x will cause deformity with this model.Flux:This is a whole other beast, with it's own rules, what language do we use here? =w=If you answered Natural, then you're a smart cutie.Flux you can either upscale or use Adetailer. not both at the same time. Flux could be slower, doesn't give the cleanest of results, but it's your go to for realistic gens. It may me a teeny weency more energy consuming than other types of checkpointsI'll be using a different set of lora to test it. Loras I've been using previously do not support Flux.FLUX.1: Decent alone, but very good with Lora, stable.HUBG_Flux.1丨Beautiful Realistic:I don't really use Flux, soooo if anyone has any good recommendations please recommend them ദ്ദി(ᵔᗜᵔ)Just Like last time ( •̯́ ₃ •̯̀) I may, need to continue you it in another part.I didn't expect that I'll have so many checkpoint recommendations.I Hope this helped you find the checkpoint for you ♡If you have any extra info to add it, or a question feel free to comment it ⸜(｡˃ ᵕ ˂ )⸝♡ I'll help as much as I can.

Welcome To Tensor! So you just started your journey in AI generation, now what?This Guide is what I wished I've known when I started Tensor. I'll share everything I've learned so far (•̀ᴗ•́ )وThis Guide will not give you a detailed explanation on how AI work, it's meant to help you get accustomed to Tensor and with generating art.Table Of ContentCommon TerminologyCreate Tool (Text2Image)Generation basicsPrompts In DepthRecommendationsCreations settingsCheckpointsLoraEmbeddingArtist TagsAdvanced Generation tipsOther Create ToolsExploring TensorAdviceTerminologyCheckpoint: It's the main model that will be used to draw. Consider this your cake. it's the base of your creation.LoRA (Low-Rank Adaptation): This is your topping, your decoration, that will distinguish your cakes from the rest.Embedding: They help the generation either focus on certain elements, or avoid deformity. consider it the oven that automatically turns off when your cake is baked, ensuring it doesn't burn. consider it a cheat sheet for the AI.SettingsSampler: How the AI denoises the image. In cake terms, the AI starts with all the ingredients, the sampler is the steps it takes to mix and bake the cake you desire.Schedulers: control how fast and in what order the AI removes noise (the steps it takes to bake)Sampling steps: The number of times the AI "refines" the image from pure noise to your final result. higher steps = better results (will discuss it more in recommendations)CFG Scale (Classifier-Free Guidance): How strictly the AI follows your text prompt vs. getting "creative."Seed: A number controlling randomness. Use the same seed if you want similar results.Clip Skip: CLIP is the part of Stable Diffusion that converts your text prompt into numerical data the AI understands, 1: means it uses all layers, the higher the value the more layers the AI will skip, causing more abstract creative generations.UpscaleIncreases the resolution of your creation, can sometimes help fix deformitiesNot always necessary, lower upscalers like 1x can sometimes make the image look muddy, if you use LoraComparsion of upscaler results using 4X_foolhardy_remacriAdetailerClean up the generation, mostly fixing deformity in the face and hands.Even if there's no deformity, it can be used to improve the quality of the image aesthaticallyUsing the face_yolov8m.pt. The ruffles of the skirt were slightly tweaked, and the face was changed (if it's an improvement or not depends on your taste)Layer DiffusionTo make a transparent backgroundNOW ENOUGH TERMS Lets go to the fun stuff!Create Tool (Text2image)Prompt: What you want the AI to generate [Figure 1]Negative prompt: what you want the AI to avoid generating. (it's not really a necessary to write in it) [Figure 2]A1111: I'm still not fully sure, something about stable diffusion, and cloud. what you need to know if you use a lot of emphasis "(), ():1.4 etc" you will be asked to either tweak your prompt or tick the A1111 [Figure 3]Translation: Translate your prompt to English (quite self explanatory) [Figure 4]Random: Generates you a random prompt [Figure 5]Abstract (VERY IMPORTANT): Verbalizes a picture you share with it. with Booru tags, or Natural Language [Figure 6]Prompt Enhance: Enhances your prompt, it removes artist prompts, some of the positive prompts. Tread carefully with using it. [Figure 7]Reset: Resetting all settings back to default [Figure 8]Presets: Saving settings, checkpoints, Lora, Prompt, negative, everything. in-case you want to replicate it, [Figure 9]Generation Basics [Let's begin with the fun! •̀ω•́ ]Pick a Checkpoint (I'll recommend ones in the the next part =w=)Loras for style, specific character, details, poses, props etc.Embedding (optional) to reduce deformityNegative Prompt (optional doesn't always work)The negative prompt I use "score_4, score_5, score_6, modern, recent, old, oldest, cartoon, graphic, text, painting, crayon, graphite, abstract, glitch, deformed, mutated, ugly, disfigured, sketch, comic, meme, poorly drawn, simple background, watercolor, lowres, low detail, text, blurry, realistic, 3d, bad anatomy, bad proportions, deformed anatomy, deformed face, deformed eyes, monochome, grayscale, text, twitter_username, artist_twitter, anatomically incorrect hands missing fingers, extra digits, fewer digits, bad eye, shota, censor, (multiple fingers), (blurry eyes), (extra legs), conjoined,ai-generated, crop out, white line, stubble"Most checkpoints have a recommendation that you can apply, which includes the negative.Prompt (The fun part =w=)Wording Anime Vs RealismIf your gonna use realistic or FLUX models: You'll have to describe the images like you'll describe a photograph, using natural language.'Full body photograph of a man with black hair sitting on bed resting his arms behind his head."For Anime [my specialty ( • ̀ω•́ )] and semi-realism models: Use Booru tags (using trigger words.)"1boy, Black hair, sitting, ,arms_above_head, on bed"Cheat Sheet =w=: Use the abstract tool to help you verbalize a picture, granted you have to provide the picture, let it be a picture you generated and would like to replicate, or an inspiration (You can only use this tool for SFW pictures only) Example Of How Abstract words my generationMy prompt and resulting pictureNow how abstract worded it: Prompt In depthFormatIf you are gonna connect two words you can do it as. "bowl_cut" instead of "bowl cut" it sometimes help the AI understand the prompt better.When generating smut (Don't worry I don't judge~) Write "rating_explicit, NSFW, Uncensored" at the start of the prompt, it'll help combat any censorship the AI will try.Separate paragraphs by using "BREAK," (not really necessary IMO)doesn't work with A1111EmphasisEvery word in the prompt has a value of 1, if you want to emphasize a word you either use "()", or "(:1.4)", and if you want to de-emphasize just do (blue eyes:0.9), any value less than 1.Every "()" increases the emphasis by 10% to a max of 30% "((()))" so if you want to emphasize blue eyes either use "(((blue eyes)))" or "(blue eyes:1.3)"if you want to use "()" without emphasizing just write it as "\(blue eyes\)""[ ]"Use it when the AI doesn't generate the thing you want. Let it be making the subject a girl instead of a boy. making their shirt a different color than the one you wrote, however emphasizing too much can cause the AI to neglect other aspects, which may result in deformity.Artist PromptWriting an artist name in your prompt can alter your creation. You can write it as(artist:rinotuna), or as rinotunaSome artists the AI doesn't recognize. You'll have to use Lora to for their style.tip If you are writing a bunch of artists, just like In the figure I showed previously regarding abstract tool, it'll be good to lower their values, so they don't clash.Here's an example of how adding an artist name in the prompt can change the style.Positive PromptPositive keywords, used to improve the end result.Like "Masterpiece, Highres, Absurdres, best aesthatic, high resolution, amazing quality, 4k, hd, ,very aesthetic, volumetric lighting, perfect lighting, detailed eyes, perfect anatomy, perfect proportions, high definition, best quality, very awa, newest, extremely detailed, highres,detailed beautiful face and eyes,best aesthetic, scenery, good colors, maximalism, (Detailed background), Dynamic lighting, deep shadows, dynamic shadows, countershading, depth of field, ultra-detailed,ambient occlusion, raytracing, HD eyes"it's unnecessary from testing, I've noticed that adding them, doesn't change the outcome of the gen.RecommendationsCreation Tool (foundation, it's boring but necessary)Preferred Aspect Ratios beside the recommended (square, landscape, portrait)512, 768 - 521, 1536 - 710, 1536768, 1536 - 1024, 1536 - 1200, 1536, 1536, 1080Samplers (a is for ancestral):Euler a: balanced in speed and qualityDPM ++ 2S a Karras: More stable than DPM ++ 2M not as high in qualityDPM ++ 2M Karras: Slower, High QualitySampling Steps:Anime, Semi-realism: 20-25Realistic, Flux Realism: 30+ (30-35 is a good median)CFG, Guidance Scale:4 - 7most checkpoints have it in the recommended settingClip Skip:1-2 (Personally I go for 6)most checkpoints have it in the recommended settingUpscaler Models:4X_ultrasharp basic but reliable4X_foolhardy_remacri (personal favorite)Esragan 4X+ :The original, can have better results than the improved versionR-Esragan 4X+ : Improved version of Esragan best for Realistic PhotosR-Esragan 4X+ Anime6B: Best For Anime, Cartoon-ish gensHere is an Upscaler models comparison:Upscaler SettingsSteps 20-30 don't make it more than your sampling stepsDenoising strength 0.4 - 0.5 feel free to test it out and find the ones you likeDenoising means there will be redrawing. It may cause deformity especially with 1.5x+if you want to higher res your picture without that much denoising, set the strength to 0.1Here's an example with 1.5X Upscale0.1 denoising strength improved the quality of the picture but didn't redraw much over it.0.4 denoising strength improved the quality of the picture more significantly, however it also caused deformity, the eyes quality improved but now the hands seem to have an extra finger.Adetailer Models:face_yolov8s.ptface_yolov8m.ptAdetailer Settings:Steps 20-25Detection model:0.5 is a perfect medianlow value=high range of repair / High value= High Repair AccuracyIt depends on your image, If there's plenty of deformity across the picture then use 0.3-0.5If the deformity and mostly in a specific place, like the face. Then use 0.5-0.75Denoising Strength: 0.4-0.5 (I personally put it 0.25 if I want that unfinished look to the image)Inpain mask blur: it's for the inpaint tool (Which I don't play with, I'll update after playing with it ദ്ദി ˉ͈̀꒳ˉ͈́ ))WITH THAT we are done with technical Jargon now for the fun part ⸜(｡˃ ᵕ ˂ )⸝♡~In the next part [Sorry but I've reached the max letter count (╥﹏╥)]If you have any question, feel free to comment itヾ( ˃ᴗ˂ )◞ • *✰

This is a guide for beginners it's very very basic, I'll showcase emotions and how the AI portrays them (or doesn't)The style I'll be using to test is Anime. The seed of the neutral expression would be used as a base and only the " neutral expression" of the prompt would be changed.Prompt: masterpiece, best quality, aesthetic, , 4k, hd, amazing quality, very aesthetic, volumetric lighting, perfect lighting, detailed eyes, perfect anatomy, perfect proportions, high definition, masterpiece, best quality, very awa, newest, highres, absurdres, year 2024, extremely detailed, highres,detailed beautiful face, high resolution, good colors, bright skin, Dynamic lighting, countershading, depth of field, ambient occlusion, raytracing, bara, face up close, face, eyes, nose, neck, chest, toned neck, adam apple, looking at viewer, head_tilt, slice-of-life style, focus face, male_only, neutral expression1boy, toned boy. handsome, ikemen, dynamic angle,, multi color hair, blue school_uniform, idol costume, (Glittering background), outdoor, school, blue sky, heart, hair ribbon, head rest, blue glitter hair streaks,shoulder-length_hair, Black hair, cyan eyes, warm-ivory_skin.Expressions: Neutral, Happy, Angry, Sad, Pout, BoredSmug, Naughty, Grimace (don't ask), Excited, Hopeful, JoyfulNervous, Pensive, Relaxed, Shy, Shocked, SleepyWorried, Mischievous, Embarrassed, Seductive, Inawe, AhegaoLastly. Emotionless, Scared, Disgust, At the bottom is an example of how to enhance the emotion, is to describe the expression through tags.I did the Smug expression but added more details of how it should look. Search for Inspo learn from other creators, check their prompts, try to verbalize a picture's expression, it'll help your gens look more lively. Emotions that didn't show up: Jealous, hurt, proud, suspicious, thoughtful, stressed,  tired, troubled, wary, embarrassed. Romantic, frustrated, fearful  confused , curious, troubled, worried, sulking, melancholic, disappointed, disgusted