Quick Generate
Select a style, write what you want to see and click Generate
Masterpiece, best quality, (tiny Hakos Baelz, full body, chibi style:1.2), hololive, mouse ears, blue eyes, short hair, twintails, tiny human, miniature, on a table, sitting at a tiny table, (american diner, 70s diner restaurant), next to a tiny pizza, ((pepperoni pizza)) detailed background, depth of field, rimlighting, specular highlights, bloom, atmospheric lighting, (colorful, vivid, sunlight, warm lighting:1.2) <lora:allVtubersLora_hll31:1>
Masterpiece, best quality, gold trim, eldritch abomination, cosmic inferno, incomprehensible, horror \(theme\), ((angel of greed), from below, close-up, monster:1.3), hyperrealistic, detailed background, depth of field, rimlighting, specular highlights, atmospheric lighting <lora:darkKukiv1:0.1>
What you want in the image.
Clear
Random

Model Description by Creator

Introduction

Isopropanol Addiction Mix (so named because of my ongoing struggles with drinking rubbing alcohol) is a semi-realistic merge of a few different models, mostly anime, that has resulted in a nice versatile model. It works surprisingly well with a lot of LoRAs, and produces very good close-up details. It is based, ultimately, on Anything v4.5 and BasilMix.

Recommended Settings

These are the settings that I use personally. Feel free to experiment.

  • Sampler: DPM++ 2M Karras

  • CFG Scale: 7.5

  • Steps: 20

  • Clip Skip: 2

  • Hires Upscale (best enjoyed with this): Either Latent (Nearest-Exact) or whatever your preferred upscaler is, such as 4x-UltraSharp. The former should have a higher denoising strength (0.5 is the hard minimum, sometimes a bit higher than that is needed, I like 0.6), and the latter can have any denoising strength, but I recommend 0.3-0.4 for it. Again, up to you. If your VRAM keeps getting gobbled up, try this extension.

  • VAE: kl-f8-anime2.ckpt, also known as the Waifu Diffusion VAE, is good and my personal choice. You can also use blessed2.vae.pt.

  • Positive prompt: See the examples above. For a more realistic look, try adding "hyperrealistic" to your prompt. Equally "anime" may shift you in the other direction, but only so far.

  • Negative prompt: A lot of models seem to have a bit of an issue with earrings and ear piercings being disproportionately common. Try adding "(ear piercing, earrings:1.2)" or similar. Beyond this, I recommend EasyNegative and bad-image-v2-39000. Try the negatives I use in the example images!

  • How Do I Get Darker Images Please Help: Behold! DarkKukiv1.safetensors at 0.3-0.4 LoRA strength. At least, that one is my favourite. You can try other offset noise LoRAs if you want.

Feel free to share what it creates in a review.

Images Generated With This Model

Masterpiece, best quality, (tiny Hakos Baelz, full body, chibi style:1.2), hololive, mouse ears, blue eyes, short hair, twintails, tiny human, miniature, on a table, sitting at a tiny table, (american diner, 70s diner restaurant), next to a tiny pizza, ((pepperoni pizza)) detailed background, depth of field, rimlighting, specular highlights, bloom, atmospheric lighting, (colorful, vivid, sunlight, warm lighting:1.2) <lora:allVtubersLora_hll31:1>
N/A
Hidden For Public Users
Masterpiece, best quality, (tiny Hakos Baelz, full body, chibi style:1.2), hololive, mouse ears, blue eyes, short hair, twintails, tiny human, miniature, on a table, sitting at a tiny table, (american diner, 70s diner restaurant), next to a tiny pizza, ((pepperoni pizza)) detailed background, depth of field, rimlighting, specular highlights, bloom, atmospheric lighting, (colorful, vivid, sunlight, warm lighting:1.2) <lora:allVtubersLora_hll31:1>
Masterpiece, best quality, gold trim, eldritch abomination, cosmic inferno, incomprehensible, horror \(theme\), ((angel of greed), from below, close-up, monster:1.3), hyperrealistic, detailed background, depth of field, rimlighting, specular highlights, atmospheric lighting <lora:darkKukiv1:0.1>
N/A
Hidden For Public Users
Masterpiece, best quality, gold trim, eldritch abomination, cosmic inferno, incomprehensible, horror \(theme\), ((angel of greed), from below, close-up, monster:1.3), hyperrealistic, detailed background, depth of field, rimlighting, specular highlights, atmospheric lighting <lora:darkKukiv1:0.1>
Masterpiece, best quality, 1man, male, close-up, (man, long beard, old man), solo, upper body, (tan, dark skin, turban, small round glasses:1.4), white open collared shirt, serious, furrowed brow, looking away, (dark, dimly lit:1.2), jungle, trees, ferns, bushes, flowers, vines, night, moonlight, blue lighting, dark robes, gold trim, hyperrealistic, detailed background, depth of field, rimlighting, specular highlights, bloom, atmospheric lighting <lora:darkKukiv1:0.3>
N/A
Hidden For Public Users
Masterpiece, best quality, 1man, male, close-up, (man, long beard, old man), solo, upper body, (tan, dark skin, turban, small round glasses:1.4), white open collared shirt, serious, furrowed brow, looking away, (dark, dimly lit:1.2), jungle, trees, ferns, bushes, flowers, vines, night, moonlight, blue lighting, dark robes, gold trim, hyperrealistic, detailed background, depth of field, rimlighting, specular highlights, bloom, atmospheric lighting <lora:darkKukiv1:0.3>
Masterpiece, best quality, 1girl, solo, from behind, backpack, winter clothes, sunset, arctic, glacier, snow, alpine forest, mountainous horizon, sky, trees, warm lighting, hyperrealistic, detailed background, depth of field, rimlighting, specular highlights, bloom, atmospheric lighting
N/A
Hidden For Public Users
Masterpiece, best quality, 1girl, solo, from behind, backpack, winter clothes, sunset, arctic, glacier, snow, alpine forest, mountainous horizon, sky, trees, warm lighting, hyperrealistic, detailed background, depth of field, rimlighting, specular highlights, bloom, atmospheric lighting
Masterpiece, best quality, no humans, castle garden, trees, summer, flowers, bushes, roses, clouds, sky, stone path, big tree, bees, butterflies, hyperrealistic, detailed background, depth of field, rimlighting, specular highlights, bloom, atmospheric lighting
N/A
Hidden For Public Users
Masterpiece, best quality, no humans, castle garden, trees, summer, flowers, bushes, roses, clouds, sky, stone path, big tree, bees, butterflies, hyperrealistic, detailed background, depth of field, rimlighting, specular highlights, bloom, atmospheric lighting
Masterpiece, best quality, 1man, male, solo, crying, tears streaming, face, on stomach, on ground, extreme close-up, looking away, cracked skin, turning to dust, mechanical limbs, glowing eyes, black armour, close-up, cyberpunk city street, pavement, neon lights, rain, reflections, puddles, trees, flowers, forest, bushes, hyperrealistic, detailed background, depth of field, rimlighting, specular highlights, bloom, atmospheric lighting
N/A
Hidden For Public Users
Masterpiece, best quality, 1man, male, solo, crying, tears streaming, face, on stomach, on ground, extreme close-up, looking away, cracked skin, turning to dust, mechanical limbs, glowing eyes, black armour, close-up, cyberpunk city street, pavement, neon lights, rain, reflections, puddles, trees, flowers, forest, bushes, hyperrealistic, detailed background, depth of field, rimlighting, specular highlights, bloom, atmospheric lighting
Masterpiece, best quality, 1girl, solo, tank top, neon trim, track pants, multicolored hair, multicolored clothes, blue eyes, smile, teeth, looking away, neon graffiti, simple artstyle, thick lineart, warm colours, cel shading <lora:darkKukiv1:0.3>
N/A
Hidden For Public Users
Masterpiece, best quality, 1girl, solo, tank top, neon trim, track pants, multicolored hair, multicolored clothes, blue eyes, smile, teeth, looking away, neon graffiti, simple artstyle, thick lineart, warm colours, cel shading <lora:darkKukiv1:0.3>
Masterpiece, best quality, 1girl, solo, orange hair, green eyes, close-up, pointy ears, green turtleneck sweater, in a forest, trees, flowers, holding a white bottle of isopropanol, detailed background, depth of field, rimlighting, specular highlights, bloom, atmospheric lighting, (colorful, vivid, sunlight, warm lighting:1.2) <lora:allVtubersLora_hll31:1>
N/A
Hidden For Public Users
Masterpiece, best quality, 1girl, solo, orange hair, green eyes, close-up, pointy ears, green turtleneck sweater, in a forest, trees, flowers, holding a white bottle of isopropanol, detailed background, depth of field, rimlighting, specular highlights, bloom, atmospheric lighting, (colorful, vivid, sunlight, warm lighting:1.2) <lora:allVtubersLora_hll31:1>
Masterpiece, best quality, 1girl, solo, ganyu \(genshin impact\), :D, mouth open, smile, closeup, headshot, 1girl, solo, forest, trees, sunlight, grass, flowers, bees, birds, hyperrealistic, detailed background, depth of field, rimlighting, specular highlights, bloom, atmospheric lighting  <lora:ganyuGenshinImpact_v10:1>
N/A
Hidden For Public Users
Masterpiece, best quality, 1girl, solo, ganyu \(genshin impact\), :D, mouth open, smile, closeup, headshot, 1girl, solo, forest, trees, sunlight, grass, flowers, bees, birds, hyperrealistic, detailed background, depth of field, rimlighting, specular highlights, bloom, atmospheric lighting <lora:ganyuGenshinImpact_v10:1>
Masterpiece, best quality, delicious burger, succulent burger, restaurant background, hyperrealistic food photograph, no humans, detailed background, depth of field, rimlighting, specular highlights, bloom, atmospheric lighting
N/A
Hidden For Public Users
Masterpiece, best quality, delicious burger, succulent burger, restaurant background, hyperrealistic food photograph, no humans, detailed background, depth of field, rimlighting, specular highlights, bloom, atmospheric lighting

Train Models

Train AI Models On Your Own Pictures

You can use your own pictures to train AI models. Reimagine yourself in new ways with our AI-powered image generator.
Create pictures of yourself around the world, in your dream outfit, as a video game character, or in countless art styles. The possibilities are endless.
Train Your Own Model

Generate AI Images. Simple and Easy

AIEasyPic is the #1 AI image generator. Generate images with simple prompts, swap faces, train models to reimagine yourself, and more.