All Images Here Are Generated By AI.

All Models

>

Soul of Disco Diffusion

Generate AI Images. Simple and Easy

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

Model Description by Creator

§ The ultimate coherent Disco Diffusion Lora §

A coherent Disco Diffusion Lora trained on 168 high quality DD renders in 768x768 from my own render library from last year. It was trained on a low rate with high dimensions (128).

§ Effect

  • Sharpness!

  • It gives many details and tends to "place" the subject in a scene (landscape or else, but only on higher weights < 0.6)

  • Very ornamental, carvings and flowy details!

  • Creates depth and strong contrast and sharp lines

  • Tends to lean into 3D render style

  • I've trained it so to retain coherence from SD by being very picky with my samples as well as let the training take it's time to learn the style.

  • Coherence of SD but style and sharpness of tiny details of high quality DD renders

§ Prompts

  • Test images on here are rendered with:
    - Realistic Vision 3

    - Chillout

    - RevAnimated 1.0

    - My own (unreleased) merge

    - CardosAnimated 1.0

You need to specify "in OliDisco style" to get it to max itself out. It is also very strong, so usually around 0.4 in strength works good.

Here are some test prompts you can check out. I didn't specify my own since I don't use the Chunk PNG feature. So you might find it hard to replicate my renders 1:1, but just ask and I'll send it your way and explain the other settings.

~~~~

in OliDisco style, 1990s_\(style\), (freckles:0.72), (skin texture:0.72), (wrinkles:1.14), skindentation, lips, reflection in eyes, detailed iris, detailed eyes, golden hair, fly_aways, messy hair, hairstrands, windy, realistic, photo, hyperrealistic, bokeh, studio, photo, retro, retrofuturistic, 1990s_\(style\), 90s vibe, 1980s_\(style\), 80s vibe, bright portrait of a victorian lady in electrical portals,surreal, ambient occlusion, global illumination, lensflare, bokeh, focused, Asher Brown Durand and Omnissiah, cosmos, maximalist, zbrush, particles, atomic, sun, moon, octane render, realistic, insane details, ellen jewett, peter mohrbacher, intricate, ornate, vibrant colors, cinematic lighting, micro details, trending on artstation, cgsociety, HDR, HD wallpaper, 8k, 3D, artistic illustration, masterpiece and beautiful fractal arts, psychedelic explorations, vivacious comics and Bougainville (patterns:0.72), fantasy art, subsurface scattering, contrast, high_contrast, art, two tone lighting, sharp focus, teal hue, octane, unreal, dimly lit, low key

Negative: cartoon, drawing, naked, cartoon, 2d, (low quality, worst quality, medium quality:1.35), (signature, typography, text, signed, autograph, scribble in the corner, website, watermark, watermarked:1.35), getty, stockphoto, stock, logo

~~~~

in OliDisco style, (Intricate photo:1.4), award for best visual effects, (masterpiece, best quality, realistic:1.1), bokeh, anamorphic, landscape, fantastic dark fantasy landscape at night, foggy, dark silhouettes, dynamic dutch angle, dynamic camera, maximalist composition, big messy unkempt, 8K, 3D, fractal, Esoteric, recursivness patterned, renderman, sacred geometry, minute, hyper-real, reality, realistic, meticulous, extremely detailed, very detailed, deep dramatic shadows, strong ambient occlusion, realistic grimy texture properties, beautiful dancing shadows, by darius kawasaki, analog, analog style

Negative prompt: paper, character, animation, animated, disney, pixar, concept, clay, game, gaming, anime, japanese, chinese, painting, cartoon, toon, framing, framed, minimalism, broche, mask, sculpture, lamp, studio, statue, bust, product, fan, woman, nude, panties, pussy, anorexic, sickly, sick, disease, tight, cross-eyed, anime, skinny, naked, stripper, sexy, manga, holes, canyons, wrigly, horror, disgusting, slime, drip, drool, grool, split, strip, divided, two pages, panel, panels, www, --bad-quality, --np_simple_negatives, --bad-picture-chill-75v, --deepNegative, (scribble, text, website, watermark, type in corner, logo:1.2), cgi, fake, render, painting, illustration, B&W

~~~~

§ Thank you for checking it out! :)

Try Full Demo
Quick Generate
Image quality might not be as good as what you can achieve with our full generation tool.
Placeholder
Choose A Style, Type What You Want In The Image And Click Generate
Placeholder
Choose A Style, Type What You Want In The Image And Click Generate
What you want in the image.
Clear
Random
* This model is available for use and will generate similar character to the ones seen in the images.
Train Your Own Model

Images Generated With This Model

Loading...

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.