Diffus
Get started
models/The Araminta Experiment (SDXL+Flux) - Fv5

The Araminta Experiment (SDXL+Flux) - Fv5

aramintastudio
|
12/4/2025
|
11:25:28 AM
| Discussion|
0
Black and white giraffe stands contemplating a gigantic dandelion on a snowy hillside under a windy sky dispersing dandelion seeds in monochromatic tones.

Recommended Negative Prompts

big boobs, ((watermark)), censored, low-res, low quality, dull, overcooked, artefacts, JPEG artefacts, poor quality, deformed, missing limb, extra limb

Recommended Parameters

samplers

DPM++ 2M SDE Karras, DPM++ 3M SDE Karras, DPM++ 2M SDE Exponential, DPM++ SDE Karras, Euler Ancestral, Euler, DPM++ 2M beta, sgm_uniform, DDEIS, normal, simple, beta, dpmpp_3m_sde, dpmpp_3m_sde_gpu

steps

12 - 25

cfg

1.5 - 11

Tips

Use DPM++ 2/3M SDE with Karras or Exponential samplers and 25+ steps with CFG around 5-7 for best SDXL results.

Try DPM++ SDE Karras with fewer steps (e.g. 12) and higher CFG (8-11) for alternative stylings.

Euler Ancestral / Normal samplers produce less detailed results that may be desirable with model Fv6 to reduce noise.

Default CLIP Skip is 2, but trying 1 can enhance prompt adherence, and 3-4 can improve concept focus.

For Flux models, samplers like DPM++ 2M beta or sgm_uniform are preferred.

CFG settings greatly influence image quality; lower CFG (1.5-2.5) avoids plastic skin effect in photos, while illustration styles may require higher CFG (3-6+).

Inappropriate CFG or steps often causes blurred or malformed images, and adjustments may be needed depending on style.

Version Highlights

Based on Fv4 with a merge with some previous versions of my models as well as BigLove_XL2 in order to push further realism.

The result is that Fv5 is my most photorealistic version so far with more precise and subtle images compared to F4 but as realism is pushed further the model is less and less able to produce stylised images, especially illustrations: a new G series version is coming soon for these uses cases :)

For the settings, DPM++ 2/3M SDE / Karras or Exponential are always a good bet with 25+ steps and CFG around 5-7. But DPM++ SDE / Karras with less steps (e.g. 12) and higher CFG (8-11) is worth a try. The default CLIP SKIP of 2 is also a good bet, but using 1 or 3-4 is also worth trying.

Creator Sponsors

If you enjoy my contribution to this community, feel free to buy me a coffee: the more caffeine I drink the more models I can create 😅

If you enjoy my contribution to this community, feel free to buy me a coffee: the more caffeine I drink the more models I can create 😅

Comparison gallery here: Fv6-Fv5 and NSFW illustration comparison between Gv1-Cv6-Fv2.

Current SOTA model in my experiment:

  • SDXL Base model: Gv4 is the most balanced model allowing both realistic and styled NSFW and SFW images. Better aesthetic than Fv6 but less photorealistic.

  • SDXL photorealist (SFW and NSFW) model: Fv6 is the way to go for hyper-realism including realistic NSFW images but it mostly lacks the styling capabilities of Gv4.

  • SDXL Illustration : Gv4 (SFW and NSFW). Cv6 is however still worth a try if you are not into NSFW images.

  • Flux model: Flux1-A1

Image Generation Settings for SDXL models

DPM++ 2/3M SDE / Karras or Exponential are always a good bet with 25+ steps and CFG around 5-7. But DPM++ SDE / Karras with less steps (e.g. 12) and higher CFG (8-11) is worth a try as is Euler Ancestral / Normal for a less detailed result which may also be interesting with a model like Fv6 to get something less "noisy".

The default CLIP Skip of 2 is also a good bet, but using 1 or 3-4 is also worth trying: 1 push more towards prompt adherence and 3-4 give sometimes a better result than the default focusing more on the "concepts".

Image Generation Settings for Flux models

My preferred settings are DPM++ 2M / beta or sgm_uniform or DDEIS / normal for the sampler / scheduler, beta giving a bolder stronger image. For a more subtle image, Euler / simple or beta seems a good bet.

CFG seems to have a huge impact on the final image and be very sensitive even to small variations.

  • For photos, CFG should remain low (1.5-2.5) to avoid plastic skin.

  • For fine art and illustration it is more complicated because it depends on the medium. For "rough" styles (painting, watercolours etc.), CFG should stay quite low in the 1.5-2.5 range but for anime or comic style, CFG needs often to be pushed further to achieve the desired style (3-6 or more).

If the image is messy/malformed or blurred, it is often because the CFG/steps are inappropriate for this image, but it is not always easy to know whether CFG/steps must be increased or decreased (at least to me 😊).

There is for sure a lot to learn concerning Flux behaviour which is quite different than SDXL and we will need to adapt.

Workflow

All my images are uploaded with embedded ComfyUI workflow which is alas incompatible with CivitAI processing and most often prompt cannot be retrieved. You can however download the original PNG image with the workflow included by clicking on the "DOWNLOAD" icon in the image viewer.

Why I try to mostly publish images straight from my model with maybe a bit of a Lora (mine or some detail enhancer), I also sometimes now use Controlnet to get better more detailed compositions more easily: in this case the source image is obviously not in the workflow, but I guess you can still use the image I publish as a source if you want to make a variation :)

Past

Starting from E serie, models are evolving sometimes by merging with other models (thanks to other contributors!), but mostly via training on my own dataset: a modest dataset (~2000 images currently), but I try to somehow compensate with quality and originality.

Starting with Fv1, I have included many synthetic images I created using previous versions: playing hard with prompt and retouching when necessary the result in Photoshop in order to have a dataset which contains many original images.

The core idea behind this model was to create a versatile tool by merging some of the best existing models which fits my personal taste (photography and fantasy art to make it simple). My primary goals were:

  1. Photorealism: The ability to produce stunningly realistic images of both people and objects/nature.

  2. Flexibility: The ability to create highly stylized images, allowing for artistic expression through various styles and combinations of artists. I am from a older generation and comes from Europe so "style" does not mean for me "Japanese kawaii aime with boobs" or "DC Comics cartoon with lots of superheroes and voluptuous blonde babes" but more from a univers of Frank Frazetta, Milo Manara, Boris Vallejo, H.R.Giger, Wojtek Siudmak and such fantasy art masters: there are boobs involved for sure, but the style is somewhat different :P

  3. As I doesn't like being limited in my exploration of the human body, the idea is also to have a fairly capable NSFW model. However, dur to the nature of available training images in the data sets, NSFW often comes with a strong bias toward either porn photos or porn Japanese anime and impact the flexibility (typically as soon as you use the word "sexy" in your prompt you need to weight in the style). This point is thus NOT the priority for the base model but is pushed forward in the NSFW model.

Previous
ICBINP XL - v4
Next
MonsterCoffeeBang! Mix - v1.1 No VAE
Use this model

Model Details

Model type

Checkpoint

Base model

SDXL 1.0

Model version

Fv5

Model hash

17b3e65f07

Creator

aramintastudio's avatar
aramintastudio

Discussion

Please log in to leave a comment.

Images by The Araminta Experiment (SDXL+Flux) - Fv5

Black and white giraffe stands contemplating a gigantic dandelion on a snowy hillside under a windy sky dispersing dandelion seeds in monochromatic tones.

base model Images

Photorealistic scene of undead characters including zombies and skeletons walking through a spooky cemetery illuminated by glowing jack-o'-lanterns under a dark, ominous sky.

nsfw Images

A detailed digital portrait of a female warrior with long black hair, red eyes, and freckles, holding a katana over her shoulder against a blue background.
A poised young woman wearing a vibrant red jumpsuit with a black collar detail, sitting elegantly on a crimson upholstered wooden chair, with straight dark hair and a detailed face.
Profile view of a cyberpunk female warrior with long pink braided hair, tattoos on her arm, wearing teal armor vest and holding a katana, set against a neon-lit industrial background.
Anime-style illustration of a blonde-haired girl with blue eyes wearing a red bikini, posing confidently at the beach with water droplets on her skin and waves in the background.
A detailed digital painting of a fierce female warrior with long red hair crouching on a stone pedestal in a graveyard, holding a sword on her back. She wears dark armor and a teal cloak, illuminated by mystical lightning in the night sky.
A young woman wearing a straw hat and dress stands with her bicycle on a country path surrounded by wildflowers, watching the sunset over a village and rolling hills.
Close-up portrait of a Viking warrior princess with a white mohawk, blue eyes, intricate jewelry, fur clothing, and tattoos, set against a night sky with nebula and stars.
A steampunk female with braided hair and goggles on her head, featuring a transparent clockwork mechanical torso with visible gears and pistons, posed confidently in a mechanical cityscape.
Close-up view of a petite woman wearing sunglasses and a leather jacket, illuminated by neon lights in a futuristic city at night with retro 1980s synthwave style.
A cyberpunk female cyborg with red oni horns and black hair, wearing a black mask and mechanical blue arms, holding a glowing blue katana sword in a dark futuristic corridor.

photorealistic Images

Silhouette of a cybernetic man with wings and long slender legs filled with photorealistic mixed media collage of urban landscapes and buildings under cinematic lighting.
Black and white futuristic image featuring metallic lines extending from top, with red-tinted foliage framing the scene and a blue pulsating diamond in center.
Photorealistic image of a maid robot lying on the forest floor with foliage intertwined in her hair and robotic body parts, dressed in a white maid dress.
Fantasy supermodel with ethereal beauty wearing a vibrant, colorful avant-garde dress, adorned with intricate jewelry, set against a swirling, vivid multicolored background.
Young female barista standing behind a rustic coffee shop counter wearing a black button-up shirt and black apron, smiling softly with natural makeup and tied-back hair.
Three women wearing nautical style sweaters and slacks standing together on a luxury yacht deck at night with a Caribbean ocean backdrop.
Photorealistic scene of a young Japanese woman in a flowing colorful sundress walking along a forest path surrounded by glowing, floating jellyfish under natural lighting.
A cute girl sitting comfortably in a cozy library, studying by candlelight with warm mood lighting and an open book in front of her.
A detailed 2D vector illustration of a woman adorned with black orchid flowers and intricate floral patterns in a baroque style.
Close-up of a young man with a short blonde goatee stubble and glasses, half-lit by morning sun, with a tabby cat resting on his shoulder.

realistic Images

Detailed realistic portrait of a young woman with beautiful blue eyes, white hair adorned with flowers and gold ornaments, highlighted by sidelighting and featuring lustrous skin.
The One Ring glowing with golden runes rests on a detailed Middle-earth fantasy map, illuminated by dramatic cinematic lighting and surrounded by miniature tree shadows.
Impressionist style oil painting of a pale-skinned sultry woman with a modern black bob haircut wearing a maroon satin sleeveless dress with a black wide belt, set against a warm golden and dark textured background with chiaroscuro lighting.
A realistic, detailed portrait of a blond man holding a large magic emerald with gold engravings, wearing an ornate blue vest decorated with green gems and intricate gold jewelry.
Realistic illustration of a girl with white hair and glowing purple eyes wearing a black crop top and light blue mini skirt, sitting among pink flowers under night sky with bright sun.
Realistic view from a spaceship cockpit showing a space traveler in a helmet piloting the ship with light speed space graphics outside and Earth in the background.
A cinematic cityscape featuring tall skyscrapers illuminated under a stormy sky with volumetric light and fog, rendered in a neon color palette.
White mecha robot in detailed armor with glowing eyes, wearing a flowing black cape, standing outdoors in bright sunlight with a blurred natural background.

woman Images

Impressionist painting of a woman in denim jeans and a plaid shirt picking apples from a tree under soft cloudy light
Black and white photograph of a woman in formal attire and a wide-brimmed hat standing on a floating ice platform beside a modern chair, reflected on calm water under an overcast sky.
A female cyborg standing in a neon-lit dystopian city alley at night with glowing red eyes, wearing a biomechanical reflective suit, surrounded by rainy urban decay and wet neon reflections.
A woman with dark hair and a red headband wearing a red and white gingham mini dress and black biker boots, sitting on a concrete ledge at night against a textured urban wall with city lights in the background.
A serene Chinese woman with long black hair adorned with ornate jewelry and traditional clothing, eyes closed peacefully with a mythical creature behind her.
Slim woman in a red long form-fitting dress with slit and fringe, wearing black gloves and stockings, futuristic glasses on her head and a feather headband, set against an art deco blue background.
Close up of a woman's hazel eye showcasing detailed eyelashes, smooth skin texture, and subtle makeup.
Portrait of a redhead woman with striking blue eyes, freckles, and long flowing hair wearing a green blouse.
Back view of a woman with hair in a bun wearing a white off-shoulder dress, standing contemplatively next to a window with warm sunlight streaming into a dim room.