A practical guide to prompting Krea 2 on fal.ai — including style transfer, model selection, and real examples across photography, illustration, and concept art.
Krea 2 is a foundation image model built for creative teams. Use Krea 2 Medium for illustration and artistic styles, Large for photorealism and raw aesthetics — and lean on style transfer and the creativity parameter to dial in exactly the look you want.
Krea 2 is Krea's first foundation image model, trained from scratch to give you full creative control over the look, feel, and direction of every image you generate. It's available on fal.ai in two variants — Krea 2 Medium and Krea 2 Large — and excels at three things: aesthetic diversity, style transfer, and a tunable creativity parameter.
This guide walks through how to prompt Krea 2 effectively, when to use each model, and includes real examples with the exact prompts used.
The fastest, cheapest and most reliable way to run genAI models. 1 API, 100s of models
fal is an official API partner for Krea 2, with day-0 availability through fal's production-ready infrastructure. Krea 2 is available in two variants:
Krea 2 Medium — Smaller, faster, and more cost-efficient, with extensive post-training for stable and consistent outputs. Strongest for illustration, anime, painting, and expressive artistic styles. Starting at $0.030 per image.
Krea 2 Large — More than twice the size of Medium, with a softer post-training profile that gives outputs a rawer, more textured character. The more powerful model overall, with particular strength in photorealism and raw aesthetics like motion blur, grain, and low dynamic range. Starting at $0.060 per image.
The examples below show the prompt used, any style references applied, and the model variant.
prompt
a vase of wilting flowers
style references and strength
model
Krea 2 Medium
prompt
male cowboy riding horse through the desert, close up shot
style references and strength
model
Krea 2 Large
prompt
a male contemporary dancer focused on dancing, contrapposto pose, body contorted, wide angle, wide angle, one hand outstretched mid-dance towards the camera, his face turned away from the camera, high contrast
style references and strength
model
Krea 2 Large
prompt
people running. extremely exaggerated wide camera angle, exaggerated shapes. risograph poster. graphic illustration. dynamic composition.
style references and strength
model
Krea 2 Large
prompt
male cowboy riding horse through the desert, close up shot
style references and strength
model
Krea 2 Large
prompt
a cat jumping sideways
style references and strength
model
Krea 2 Medium
prompt
close-up macro photograph of a lower eyelid and cheek, extreme proximity revealing every pore and fine vellus hair on pale skin flushed with a soft mauve-pink bloom. The lash line dominates the upper frame — sparse, translucent lashes catching light against deep red-stained waterline. Skin texture is the subject: matte, powdery in quality, like the surface of a peach or unglazed ceramic. cakey makeup texture. The blush diffuses at its edges like watercolor into wet paper. No smoothing, no retouching — every micro-texture intact. Eyelashes coated with white mascara.
model
Krea 2 Medium
prompt
resting cheetah in close-up profile facing right, golden fur with dense black spots, distinctive black tear marks, sharp focus on amber eye and textured fur, paws on flat reddish-brown stone, blurry spotted fur in lower right, shallow depth of field, soft dark green blurred background, warm natural daylight, wildlife photography
model
Krea 2 Medium
prompt
Vivid green red-eyed tree frog, bulging bright orange eyes, vibrant orange sticky toes gripping a dark leaf, frontal macro portrait, vertical green stem on right edge, pitch black background, sharp facial focus, dramatic lighting
model
Krea 2 Medium
prompt
sweeping white parametric building, curved grid glass facade, vast light grey plaza, single figure in dark clothing walking, clear blue sky, architectural photography
model
Krea 2 Medium
prompt
A young woman laughs brightly with eyes closed against a clear, pale blue sky. Loose strands of her dark, windblown hair whip across her face. She wears a sleeveless white lace top and bright red lipstick. Shot from a low angle in soft natural daylight, isolating the figure against the minimalist sky.
model
Krea 2 Medium
prompt
A towering rust-orange suspension bridge stretches across deep blue-green water beneath a bright sky. Photographed from a low vantage point on the water, choppy waves and white sea foam churn in the immediate foreground. The bridge features massive vertical towers with intricate cross-bracing, supporting sweeping main cables and fine suspender ropes. Rolling, sunlit hills with earthy brown slopes and sparse green vegetation form the distant coastline, where tiny white sailboats glide near the rugged shore. The pale blue sky is lightly textured with scattered wispy cirrus clouds. Clear, natural daytime sunlight illuminates the vibrant orange steel structure, creating a sharp visual contrast against the muted earthy tones of the distant landscape and the dark, textured surface of the water.
model
Krea 2 Large
prompt
fantasy concept art of a dark blue skinned elven archer riding a white tiger, solid green backdrop, flat graphic design, distorted proportions and highly dynamic expressive composition
model
Krea 2 Medium
prompt
a low-angle, tightly cropped shot of a quirky, minimalist horse character straddling a simplified astronaut, depicted in a naive flat illustration style with a grainy risograph texture, using bold black outlines and a vibrant, awkward color palette of teal, orange, and yellow against the astronaut's white suit.
model
Krea 2 Medium
prompt
bold graphic poster illustration, stylized horse with an elongated fluid body and wavy serpentine tail, primitive folk art style, heavy ink texture, hand-lettered sans-serif typography reading "run wild", off-white textured paper background, risograph or linocut print aesthetic, vintage graphic design, flat 2d shapes, matte finish, high contrast, organic wobbly edges, abstract composition
model
Krea 2 Large
prompt
flat 2d vector illustration, a grey cat in a confident walking stride profile view with tail held high, background features a speeding yellow taxi cab with exaggerated horizontal motion blur and speed lines, deep blue windows, colorful traffic light on the right with pink and blue accents, solid bright chroma green background, geometric floating stars, bold black outlines, vibrant high saturation colors, kinetic pop art style, animation keyframe aesthetic
model
Krea 2 Medium
prompt
an ukiyo-e woodblock print of traffic including a cybertruck on the golden gate bridge.
model
Krea 2 Medium
prompt
colossal red torii gate on a jagged black cliff, solitary warrior silhouetted centrally beneath the archway, stone lanterns flanking the path, massive black sphere dominating a vibrant crimson sky, dark rocky foreground terrain, minimalist two-tone palette of pure black and blood red, high-contrast graphic illustration
model
Krea 2 Medium
prompt
A luminous pointillist painting of elegantly dressed figures strolling a sunlit riverbank. Men in long coats and bowler hats walk a grassy path near a woman in a white dress and resting groups. Rowboats navigate the bright blue river, backed by a stone quay and red brick buildings. The composition is rendered entirely with tiny, stippled dots of vibrant, complementary colors, creating a highly textured, radiant scene under a bright sky.
model
Krea 2 Medium
prompt
childish crayon drawing, a dragon spraying a fire extinguisher at some fire
model
Krea 2 Large
prompt
a horse in a scene from steamboat willie
model
Krea 2 Medium
prompt
sesame street style horse
model
Krea 2 Medium
prompt
A vibrant 3D animated group of five anthropomorphic animals stands against a clear blue sky. A tall skeptical giraffe in a tweed vest stands beside an eager bulldog in suspenders. A surprised squirrel in a lab coat gestures in the center. A massive smiling rhinoceros wearing a denim jacket carries a tiny mechanical crab on its shoulder. Blurred green foliage frames the bright CGI comedy scene.
model
Krea 2 Medium
prompt
Stylized 3D toy character on a solid cobalt blue background. The figure features an oversized head, large black eyes, and pink-purple hair in twin buns. It wears a textured blue sweater, green shirt, patterned blue pants, and chunky black-and-white metallic-soled sneakers. A black crossbody strap with a white buckle crosses the chest. Smooth studio lighting highlights vibrant matte textures.
model
Krea 2 Medium
prompt
dynamic angle of a model lying on the floor, wearing dark denim jeans and white socks with brown shoes, posing for an advertising campaign for Levi or Gap denim jeans. He is posed as if he is tightly trapped within the composition of the photo. while wearing a light grey long-sleeve shirt and a navy jacket. His face is clean-shaven. Photographed in the style of Tim Walker.
model
Krea 2 Medium
prompt
cinematic portrait photography, studio shot, model with closed eyes and head tilted back, wet-look black hair, glossy lips, chunky gold earring, silver chain strap, against a large defocused sphere background, dark studio, dramatic lighting, medium shot, soft focus, sensuous mood
model
Krea 2 Medium
prompt
cinematography in a film in the style of Zhang Yimou and Raise the red lantern. Moody lighting, saturated colors.
model
Krea 2 Medium
prompt
A serene woman with black and silver braids stands against a seamless white backdrop, eyes closed with her head resting gracefully on her hand. She models an avant-garde white ensemble featuring massively oversized puffed sleeves and a voluminous, asymmetrical, semi-sheer layered skirt. Pointed lilac heels with dark ankle straps complete the look. Soft high-key studio lighting accentuates the crisp, airy drapery in this minimalist fashion editorial portrait.
model
Krea 2 Medium
prompt
man with bleached buzz cut and face in deep shadow, neck tattoos, black turtleneck, thick silver chain, dramatic rim lighting, royal blue background
model
Krea 2 Medium
prompt
surreal composite photograph, a giant dark skinned man in an oversized charcoal double-breasted suit and deeply pleated wide-leg trousers striding as a towering figure over a miniature San Francisco cityscape, one worn brown suede derby shoe lifted mid-stride looming enormous in the foreground close to the camera, the tiny city spread below — recognizable San Francisco landmarks rendered in miniature including the Transamerica Pyramid, Salesforce Tower, Victorian painted ladies, and steep hilly streets with cable car tracks, tiny yellow cabs and cars dotting the roads below, black beanie, narrow dark sunglasses, woven black crossbody bag, black crow perched on shoulder wings slightly spread, figure looking coolly downward or straight ahead, bright vivid sunny blue sky, sharp photorealistic composite, wide angle lens with strong perspective distortion, cinematic scale and grandeur
model
Krea 2 Medium
prompt
A robotic head in profile reflects a city skyline and trees under the warm, harsh glare of late-afternoon golden hour sunlight. The gleaming, polished metallic surface of the artificial face captures sharp, vibrant reflections of architectural structures and dense foliage. Direct, intense, directional lighting casts deep, solid, sharp-edged shadows across the precise mechanical components and jointed paneling of the head. The aesthetic mirrors crisp medium format analog color film photography, utilizing a highly saturated, warm palette dominated by bright golden hues, stark whites, inky shadow blacks, and rich amber undertones. The scene features a hyper-realistic, densely textured rendering quality with razor-sharp focus and fine organic film grain. The high-contrast, sun-drenched exposure powerfully defines the geometric forms, metallic luster, and complex mirrored properties of the robotic profile.
model
Krea 2 Medium
prompt
A vase of oranges and two pomegranates rendered in a retro-futuristic Y2K cyber-aesthetic, characterized by hyper-reflective liquid metal and heavy airbrush textures. The central vase is constructed from elongated, razor-sharp tribal chrome tendrils and fluid, spiky cyber-sigilism shapes that interlock and sweep outward into sharp points. Nestled within the jagged metallic thorns are smooth, highly polished metallic orange spheres and two deep ruby-magenta pomegranate forms, all coated in a glassy, iridescent gloss. Blinding, multi-point white starburst lens flares erupt from the intense specular highlights across every sharp tip and curved spherical surface. The metallic fruit and spiky vessel sit against a smooth, stylized gradient backdrop transitioning from deep, saturated cobalt blue at the top to a hazy, neon cyan and magenta gradient at the base. The highly polished liquid chrome surfaces cast starkly mirrored reflections of the vivid pink and cyan environment, creating high-contrast chromatic edges. The rendering technique mimics early 2000s vintage 3D CGI artwork, saturated with a coarse, heavy film grain and a distinct digital noise texture overlaying the entire composition.
model
Krea 2 Medium
prompt
two dancers in a dramatic dip, woman in an emerald gown leaning against an art-deco lamppost supported by a man in a white tuxedo, artificial studio-set desert with painted dunes and purple cacti, theatrical painted sunset backdrop, vintage 1950s Technicolor film style, wide cinematic shot, highly saturated colors
model
Krea 2 Medium
Both models share the same parameters and features. The difference is in size, training approach, and the kinds of outputs they produce best.
Smaller, faster, and more affordable. Medium has more post-training enhancements, which makes its outputs stable and consistent across generations. It excels at illustration, anime, painting, and creative or artistic styles — a strong default for most use cases.
More than twice the size of Medium, with a softer post-training approach. This gives Large a more raw, textured, and flexible feel. It's the more powerful option — at its best, it produces results Medium simply can't match. Large shines in photorealism, expressive aesthetics, and "raw" looks like motion blur, film grain, and low dynamic range.
When to use which:
Krea 2 responds well to specific material and rendering language. Instead of "a photo of a frog," write "frontal macro portrait, vibrant orange sticky toes gripping a dark leaf, pitch black background, sharp facial focus, dramatic lighting." The more specific the visual language, the more precise the output.
Krea 2's style transfer system can extract the style of one or several reference images and apply it to your output with precision. Each reference image gets a strength value (0–100%). You can mix multiple references — Krea 2 will blend them.
The creativity parameter is unique to Krea 2. At higher values, the model adds aesthetic depth, interpretation, and visual richness beyond what you explicitly describe. Useful when you want the model to make interesting decisions rather than strictly execute your prompt.
Whether you're going for photorealism or illustration, camera language helps: "low-angle," "wide angle," "close-up macro," "contrapposto pose," "shallow depth of field." These translate directly into compositional choices.
Krea 2 is one of the most expressive image models available, built for creative teams who need real control over aesthetic output, not just prompt-to-image generation. Whether you're working with style references, dialing in the creativity parameter, or writing highly specific prompts, the examples in this guide show what's possible across photography, illustration, concept art, and beyond. Explore more models on fal.ai and find the right one for your pipeline.
Growth at fal.ai
Hundreds of production-ready image, video, and audio models behind one API.
A detailed comparison of Google's Nano Banana Pro and Nano Banana 2 image generation models.
![What is H3 Max Camera Controls (Multi-Angle) & How To Use It? [2026]](https://refinery.fal.media/url/https%3A%2F%2Fv3b.fal.media%2Ffiles%2Fb%2F0aabbf2b%2Fq8usbuG4heGtr4-AYyqfk.jpg/tr:w-1280,q-80/q8usbuG4heGtr4-AYyqfk.webp)
A guide to H3 Max Camera Controls on fal, the 3D video model that moves the camera around a still scene, covering camera trajectories, the freeze prompt, source images, and pricing.
![Seedance 2.5 Workflows: Text-to-Video, Image-to-Video, References & Editing [2026]](https://refinery.fal.media/url/https%3A%2F%2Fv3b.fal.media%2Ffiles%2Fb%2F0aa63087%2FWbpT8JAY2FG_naNgsprJl.jpg/tr:w-1280,q-80/WbpT8JAY2FG_naNgsprJl.webp)
A prompting guide for Dreamina Seedance 2.5 on fal, covering the four prompt shapes and all three endpoints, from text-to-video through image-to-video, reference-to-video, and editing.
