Model
Trends
.ai
Model
Trends
.ai
the best open source ai models

Light Slider for Krea 2: deepen shadows while the subject stays lit

Topic: Models
By Captain
Published 2026-09-30
Share
Example renders
Prompts and settings shared by the people who made these renders on Civitai. Pick a render to see how it was made.
Settings
Sampler
er_sde_simple
Steps
9
Guidance
1
Seed
537186205410151
Size
2496x3344
Clip skip
2
Weight used
1.75
Prompt
A detailed 3D digital render of a biomechanical alien entity with a glossy, obsidian-like surface texture. The creature's head occupies most of the frame, featuring large, deep black ocular cavities that reflect sharp pinpricks of light. Its facial structure is composed of intricate, flowing organic filaments and ribbed plating that transition from deep midnight blue to shimmering translucent amber. The creature possesses large, delicate fins or membranous structures extending from its head, which are semi-transparent and webbed with fine, glowing golden veins. These ethereal appendages spread out towards the edges of the frame, creating a sense of complex depth. Tiny, luminous particles and glittering dust motes float around the subject, catching the light like stars in a dark void. The lighting is highly dramatic, utilizing sharp, directional rim lighting that defines the creature's sleek contours against a pitch-black, infinite background. A shallow depth of field causes the far edges of the translucent fins to blur into soft, glowing bokeh, while the central facial details remain incredibly sharp and tactile, emphasizing the wet, reflective sheen of its extraterrestrial skin. YFG-Acrylic style Photorealism, flat colour profile cinematic flat colour grading washed tones, lifted blacks Negative: (cum, water, splashes, text, logo, watermark:-1.4) <lora:krea2_turbo_lora_rank_64_bf16:0.95> <lora:RealismSlider-v1:0.8> <lora:real_3d_krea2_loraholic:0.5> <lora:realism_engine_krea2_v3.1:1.4> <lora:detail_slider_krea2_loraholic:2.0> <lora:YFG-Acrylic-Krea2_2k-v1:1.0> <lora:light-chiaroscuro_krea2_loraholic:1.75> <lora:KREA_MIDJ_1:1.0>
Negative prompt
Negative: (cum, water, splashes, text, logo, watermark:-1.4)
Settings
Sampler
Euler
Steps
8
Guidance
1
Seed
3503669331328122
Size
2160x2880
Weight used
5
Prompt
whimsical storybook illustration,, GrimmsPaint, A whimsical hand-painted storybook illustration of a chubby elderly granny cheerfully washing an Nvidia RTX 5090 graphics card in her kitchen sink. Front view at eye level. The granny stands centered behind the sink, facing the viewer directly. She holds the graphics card horizontally in front of her with both hands: her left hand grips the left end of the card, her right hand grips the right end and scrubs the frame with one single yellow sponge under running soapy water from the faucet. Both hands are clearly visible, natural, and anatomically correct, firmly supporting the weight of the card. She has a round soft face, plump cheeks, cheerful wrinkles, deep smile lines, small crow's feet, thin-frame glasses slipping slightly down her nose, and grey hair gathered into a loose messy bun. She looks directly at the viewer with bright amused eyes and a warm cheerful expression. Her smile is broad and friendly, with neat simple visible teeth. She wears a faded floral blouse with rolled sleeves and a cream-colored apron. Soap bubbles cling lightly to her hands and forearms. The RTX 5090 is a sleek, fully enclosed rectangular block, like a solid metal case. A continuous brushed-silver metal frame wraps smoothly around all four edges, forming clean rounded corners. The front face is a flat dark graphite-black panel with one large cooling fan on the right side and large readable lettering "RTX 5090" on the left side. The front faces the viewer clearly, with the fan and the "RTX 5090" text fully visible, prominent, correctly oriented, and easy to read. White soap foam is smeared along the silver frame and around the fan housing, while the "RTX 5090" label stays clean and fully legible. The lower half of the card rests in milky dishwater inside a large stainless-steel kitchen sink. The graphics card is painted in exactly the same hand-painted storybook style as the rest of the image: the same visible brush strokes, painterly texture, softened edges, and warm illustrated lighting as the granny, the sink, and the kitchen, fully integrated into the painting. Playful moderate clusters of soap bubbles float in the sink and around the card edges, leaving the granny's face, the fan, and the "RTX 5090" lettering in clear view. The kitchen is cozy and lived-in, with pale blue subway tiles, warm wooden countertops, a sponge rack, ceramic bowls, a glass jar with cleaning sponges, a kettle, cupboards, and a softly glowing kitchen window. The whole image is a richly painted children's storybook illustration with visible brush strokes, softly simplified shapes, gently exaggerated proportions, expressive character design, painterly surfaces, hand-painted shadows, textured pigment, softly blended colors, and charming imperfections. Every object in the scene, including the graphics card, shares the same illustrated painterly style. The image is colorful, cozy, humorous, centered, and unmistakably hand-painted.
Settings
Size
1440x2160
Prompt
{a breathtaking side-profile | a provocative half-side angle | a cinematic wide shot} of a {statuesque | towering | willow-thin} secretary with an {hourglass | exaggeratedly curvaceous} silhouette and {voluptuous | ample | striking} breasts, sitting with {elegantly crossed | tightly interlaced} legs behind a {massive mahogany | ornate obsidian | gilded art-deco} desk inside a {hyper-detailed | opulent | lavishly decorated} private office at {midnight | the dead of night | a moonlit hour}. She is the picture of corporate perfection, featuring {immaculately sculpted | sleek, glossy | high-fashion} {raven black|dark|ginger} hair and {flawless, sultry | sharp, professional | high-glam} makeup, accented by {dripping diamond | ornate gold | shimmering pearl} earrings and {deep crimson | glossy black | metallic dark blue} painted fingernails. Her attire is a daring blend of power and provocation: a {sheer, translucent | see-through} top that reveals the contours of her skin, a {micro-mini | skin-tight | daringly short} skirt, {sheer black | shimmering | dark charcoal} pantyhose, and {stiletto | sky-high | pointed-toe} high heels. The setting is drenched in {moody noir | dramatic chiaroscuro | neon-tinged} lighting, with the city skyline glowing through {floor-to-ceiling | massive arched} windows in the background, while the interior is filled with {heavy velvet curtains | gold-leaf trim | intricate carved woodwork}. The camera captures the scene in {8k resolution | hyper-realistic photography | a cinematic film still}, focusing on the {tension | luxury | seductive atmosphere} of the room, with a {shallow depth of field | sharp, crisp focus} on her {confident | teasing | commanding} expression.
Settings
Sampler
er_sde
Steps
12
Guidance
1
Seed
1979114714
Size
832x1216
Prompt
Subject: A beautiful girl with long straightened hair. Deep voluptuous cleavage. Personality: Confident, slutty, well aware of her femininity, sexuality and urges. Girl appearance: Perfect slim body, narrow hips, thin waist, flat stomach, her round butt creates a stunning contrast to her tiny waist. slender form. Large round firm breasts. Harmonious facial features, fair skin. Expressive eyes. Elegant posture. long lashes, eyeliner. Smoky makeup. Pose: Interesting pose that brings the image to life. Arched back posture. Scene: In a retro diner booth under warm pendant lights, a blonde woman in a cherry-red vinyl mini skirt and white cropped tee chews a giant pink bubble of gum with her boots resting on the table, one arm draped over the back of the booth, heavy-lidded bored stare, chrome stools and checkered floor behind her, cocky 1950s-inspired defiance. neon sign says diner beautifully composed surroundings with environmental detail, layered background elements, subtle atmospheric effects, realistic textures, soft interaction between light and the environment, visually balanced but interesting composition that complements the subject without distracting from the subjects. Style: social media post, instagram, onlyfans High view angle as if viewed by a taller man. Camera: medium format, 70mm, f/2.8 <segment:face|hair,0.4,0.5//cid=11>
Negative prompt
fat, chubby, asian, indian, bare ass, exposed butt,
Settings
Size
1440x2160
Prompt
{a breathtaking side-profile | a provocative half-side angle | a cinematic wide shot} of a {statuesque | towering | willow-thin} secretary with an {hourglass | exaggeratedly curvaceous} silhouette and {voluptuous | ample | striking} breasts, sitting with {elegantly crossed | tightly interlaced} legs behind a {massive mahogany | ornate obsidian | gilded art-deco} desk inside a {hyper-detailed | opulent | lavishly decorated} private office at {midnight | the dead of night | a moonlit hour}. She is the picture of corporate perfection, featuring {immaculately sculpted | sleek, glossy | high-fashion} {raven black|dark|ginger} hair and {flawless, sultry | sharp, professional | high-glam} makeup, accented by {dripping diamond | ornate gold | shimmering pearl} earrings and {deep crimson | glossy black | metallic dark blue} painted fingernails. Her attire is a daring blend of power and provocation: a {sheer, translucent | see-through} top that reveals the contours of her skin, a {micro-mini | skin-tight | daringly short} skirt, {sheer black | shimmering | dark charcoal} pantyhose, and {stiletto | sky-high | pointed-toe} high heels. The setting is drenched in {moody noir | dramatic chiaroscuro | neon-tinged} lighting, with the city skyline glowing through {floor-to-ceiling | massive arched} windows in the background, while the interior is filled with {heavy velvet curtains | gold-leaf trim | intricate carved woodwork}. The camera captures the scene in {8k resolution | hyper-realistic photography | a cinematic film still}, focusing on the {tension | luxury | seductive atmosphere} of the room, with a {shallow depth of field | sharp, crisp focus} on her {confident | teasing | commanding} expression.
Settings
Sampler
exponential/res_2s bong_tangent
Steps
2
Guidance
1
Seed
685836848987925
Size
1880x1256
Prompt
A multi-layer dry oil painting with expressive brush strokes depicting a man in an office shirt and tie standing in front of a bathroom mirror, holding a rag in one hand and rubbing it across his face. In the reflection of a cracked mirror, he sees his reflection, the face of which is completely erased, in its place is a deep black hole from which thick black paint pours. The light in the restroom is subdued, dim, and spotlights the scene. The palette is desaturated, almost black and white. <lora:Krea2_TextFusion_Refusal_Reduction:1.0> <lora:KREA2_TA86_3xperiment0_light:0.6> <lora:light_krea2_loraholic:-1.0>

Overview

The
Light Slider
for
Krea 2
prioritizes the darkest parts of the scene: positive values deepen shadows and raise contrast while the focal areas keep some light. It is different from the author's Brightness Slider, which shifts exposure of the whole image.
The author's caution: light
LoRAs
always degrade quality a little, so use prompting as the main light control and the slider for a final push. The samples were made with a two-pass workflow where the LoRA ran only on the first pass.

Reference

Name
Type
What it is
LoRA
weight
-5 to 8
2-3 for moody scenes.
Trigger
none
Passes
first pass only
Keeps quality high in two-pass workflows.
Steps / guidance
8 / 1
Alternative
Brightness Slider
For overall exposure.
AD
No install required - run AI models on the cloud
BitVector Prism is the easiest way to start: pick a model, type a prompt and generate in a clean web app, with nothing to configure. BitVector is also available on Discord and on the web (SpyGlass).

Step by step

  1. Describe the light in the prompt first (single candle, moonlight).
  2. Add the Light Slider at 2 on the first pass.
  3. Generate and compare.
  4. Push to 4 for noir scenes.
  5. Use -2 for a flatter, overcast look.

Examples

Noir

A detective in a trench coat under a single streetlamp, fog, 35mm, black and white.
Light Slider 4

Storybook

A whimsical hand-painted storybook illustration of a cheerful granny washing a graphics card in her kitchen sink, warm window light.
Light Slider 1.5

Creative idea: candlelit still life

A pewter jug, bread and grapes on a dark table lit by one candle, Dutch painting style.
Light Slider 3

Tips

  • Low-key portraits and noir scenes are the use case.
  • Combine with the Wetness slider for rainy night streets.
  • Keep subjects lit by naming the key light.
  • Check the author's prompting guide for perspective and zoom.

Troubleshooting

Crushed blacks

Why it happens
Weight above 5.

How to fix it
3 and describe fill light.

Soft details

Why it happens
LoRA on all passes.

How to fix it
First pass only.

AD
PirateDiffusion
No install required - run AI models on the cloud
PirateDiffusion is Telegram only and built for pros: thousands of models, LoRAs and workflows driven by chat commands, with unlimited generation on a fixed price plan.

Questions

Light or Brightness?

Light for contrast and shadows; Brightness for exposure.

Trigger?

None.

Links and sources

Models in this guide

Written by
Captain

Related guides

Models
UltraReal for Krea 2 and Klein 9B: texture back into smooth AI images
vizsumit's UltraReal LoRAs reduce the soft, smooth AI look and add real texture and detail, especially for close-ups and clothing. Learn versions, weights and prompts.
By Captain
2026-10-02
Models
Krea2 Melancholy v2: cinematic, restrained, emotional realism
zidius's Melancholy LoRA for Krea 2 produces contemplative, cinematic imagery with subdued palettes, shallow depth of field and emotional restraint. Learn its zidiusArt trigger, weight and ideas for quiet scenes.
By Captain
2026-10-03
Models
Fat/Skinny Weight Slider for Krea 2: body weight as a number
Loraholic's Weight Slider adjusts body mass in Krea 2 across a -10 to 10 range designed for predictability. Learn how to use it, stability notes and ideas for character design and fitness illustration.
By Captain
2026-10-01
Models
Krea2 Enhancer: cleaner detail and lighting for any Krea 2 style
vrgamedevgirl's Krea2 Enhancer LoRA polishes Krea 2 images across realistic, anime, product and fantasy styles without changing the prompt's intent. Learn the weight, workflow and ideas.
By Captain
2026-09-30

More guides

Cloud services
Flat fee vs tokens: why unlimited plans like Graydient.ai are the best value for AI creators in 2026
A ranked cost comparison, with prices checked on 8 October 2026, of flat-fee unlimited plans like Graydient.ai against token and credit pricing from Midjourney, Leonardo, fal.ai, Replicate, Runway and Kling: what 1,000 images and 1,000 videos a month really cost on each, and why one fixed fee for unlimited images, video, audio, Grok LLM chat and web apps is the best value for creators who iterate.
By Captain
2026-10-08
Models
FLUX.1 Dev: how to prompt it, best settings and creative ideas
A practical guide to FLUX.1 Dev by Black Forest Labs: natural-language prompting, guidance and step settings, LoRA stacking, text rendering and prompt ideas that play to its strengths.
By Captain
2026-09-30
Models
Stable Diffusion XL 1.0: prompts, settings and what it still does best
How to get the most out of the official SDXL 1.0 base model: native resolutions, CFG and sampler settings, the refiner, negative prompts and ideas for styles where SDXL still shines.
By Captain
2026-10-01
Models
Stable Diffusion 1.5: the classic model, prompted properly
Stable Diffusion 1.5 is still worth knowing: the right resolution, CFG and sampler, how to use its huge library of LoRAs and embeddings, and creative prompt ideas suited to a 512-pixel model.
By Captain
2026-10-02
Models
FLUX.2 Dev: prompting the newest Black Forest Labs model
FLUX.2 Dev brings larger prompts, better text, stronger realism and multi-reference editing. This guide covers the turbo workflow, guidance and resolution settings, prompt structure and ideas that use its new strengths.
By Captain
2026-10-03
Models
Qwen-Image: long prompts, perfect text and bilingual posters
Qwen-Image is the model to reach for when the words in the picture matter. Learn how to write its long descriptive prompts, render English and Chinese text accurately, set CFG and steps, and explore layout-heavy creative ideas.
By Captain
2026-09-29