Text-to-Image
Diffusers
lora
template:diffusion-lora

Krea2 AsianMix Lora v3.0 - pre-release

Prompt
A young woman lies supine in shallow, clear ocean water over a sandy seabed. She wears white bikini top and bottoms; her long dark hair fans out underwater, strands floating gently around head. Eyes closed in serene relaxation or sleep; lips slightly parted revealing teeth — peaceful expression visible above waterline. Sunlight filters through rippling surface, casting dappled highlights on skin and sand below; soft shadows accentuate contours of torso & limbs. Water clarity reveals submerged pebbles, seaweed fragments scattered across bottom texture — natural beach environment implied by context. High-angle shot captures full body diagonally framed within composition; shallow depth of field blurs distant background while keeping subject sharply focused. Artistic style leans toward dreamy, ethereal realism with cool-toned color grading emphasizing blues and greens contrast against warm skin tones. Lighting is bright but diffused, suggesting midday sun filtered through gentle waves — no harsh shadows present. Overall mood: tranquil solitude immersed in natures embrace, evoking summer serenity and vulnerability to environment.
Prompt
The image presents a close-up portrait of a young woman with dark, wet-looking hair partially obscuring her face, her gaze fixed directly at the viewer through a lens or screen that distorts the scene with vibrant, glitchy neon effects. Her eye is the focal point, reflecting a kaleidoscopic array of pink, blue, and purple hues, as if caught in a digital mirror or holographic interface. The rest of her face is softly blurred, with only the lower half visible, her lips slightly parted and tinged with a faint pink glow. Surrounding her are abstract, floating orbs of saturated pink, cyan, and yellow light—bokeh-like artifacts that suggest a digital or dreamlike environment. The overall aesthetic is cyberpunk-inspired, with a surreal, glitch-art texture overlaying the image, giving it an otherworldly, immersive quality. The lighting is high-contrast, with sharp edges and saturated colors that emphasize the artificial, synthetic nature of the scene. The composition feels intimate yet detached, as if the viewer is peering through a digital veil into a private, surreal space.
Prompt
A young woman with dark, softly tousled hair and a warm, radiant smile stands on a train station platform, her gaze gently directed toward the camera. She wears a delicate white lace top with short sleeves and a black-and-white checkered skirt, accessorized with a woven straw bag slung over her shoulder and subtle, elegant earrings. In her hands, she holds a smartphone, fingers poised as if mid-scroll or about to tap. Behind her, a train with a red stripe and illuminated destination signs—東京 and — is stationary, its windows reflecting the stations ambient light. The platforms safety line glows faintly red, and blurred figures of commuters move in the background, adding depth and realism to the scene. The lighting is soft and natural, suggesting daytime, with diffused overhead illumination typical of urban transit hubs. The overall composition is clean, composed, and emotionally inviting, capturing a candid yet polished moment of modern urban life. The image is a realistic photograph, rich in texture and atmosphere, with no visible text beyond the trains destination indicators.
Prompt
A young East Asian woman with short, tousled brown hair sits on a rumpled white bedsheet. She wears an oversized yellow sleeveless dress with deep V-neckline, revealing black bra straps beneath her collarbone. Her gaze is direct and soft-lidded; lips slightly parted, expression calm yet intimate — not smiling but emotionally present. Skin glows under warm natural light from left, casting gentle shadows across her neck and chest fabric folds. Background is plain off-white wall — minimalist studio setting emphasizing subject texture: linen dress wrinkles, skin smoothness. Shot at eye level using shallow depth of field; background blurred to isolate figure compositionally centered, occupying lower two-thirds frame. Style: high-resolution portrait photography with cinematic color grading — muted tones except vibrant yellow dress as focal point. Mood: serene, vulnerable yet composed; evokes quiet intimacy without overt sensuality despite revealing neckline and posture leaning forward slightly toward camera.
Prompt
A lone figure stands silhouetted against a surreal, dreamlike sky, bathed in an otherworldly blend of deep indigo and radiant magenta light. The figure, seen from behind, carries a large, textured bag slung over one shoulder, their short hair catching faint glimmers of the ambient glow. Around them, particles of light drift like cosmic dust, swirling in the air as if caught in a silent, radiant storm. Jagged beams of pink energy slice through the atmosphere, intersecting with the figures form and casting an ethereal, almost ghostly aura. Below, the ground is obscured by shadow and blurred vegetation, hinting at a desolate or abandoned landscape. The entire scene pulses with a surreal, cinematic energy, evoking a sense of mystery and isolation. The image is rendered in a stylized, grainy aesthetic reminiscent of analog film or digital art, with a heavy emphasis on contrast and color saturation. No text appears in the image. The artistic style is a fusion of cyberpunk surrealism and atmospheric photography, blending realism with fantastical elements.

Trigger words

You should use asian woman to trigger the image generation.

Download model

Download them in the Files & versions tab.

License

This repository contains LoRA adapter weights trained for krea/Krea-2-Turbo, which is part of the Krea 2 model family.

Upstream Krea 2 license

Use of this LoRA requires continued compliance with the original Krea 2 Community License Agreement and Krea's applicable Acceptable Use Policy. The upstream license remains in full effect. If these additional terms conflict with the upstream Krea 2 license, the upstream license controls.

Additional non-commercial terms for this LoRA

Copyright (c) 2026 hinablue. All rights reserved except for the limited non-commercial permission granted below.

You may download, use, copy, and modify these LoRA adapter weights solely for personal, academic, research, evaluation, or other non-commercial purposes. Any redistribution or derivative version must retain this notice, clearly identify its modifications, and apply the same non-commercial restriction.

Without prior written permission from hinablue, you may not:

  • sell, rent, sublicense, or commercially redistribute these LoRA weights or derivative weights;
  • use this LoRA, directly or indirectly, in a paid product, paid service, API, hosted generation service, subscription, client work, advertising, promotion, or other revenue-generating activity;
  • use images or other outputs generated with this LoRA for commercial purposes; or
  • remove or misrepresent the attribution, upstream license, or these additional terms.

For commercial licensing permission, contact the repository owner before use.

THE LORA IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE, AND NON-INFRINGEMENT.

Downloads last month
955
Inference Providers NEW
Examples

Model tree for hinablue/Krea2-AsianMix-v3

Base model

krea/Krea-2-Raw
Adapter
(1391)
this model