
Krea 2 Large
Larger Krea 2 variant for rawer, more flexible outputs with stronger photorealism and weighted reference control
Krea 2 Large
Larger Krea 2 variant for rawer, more flexible outputs with stronger photorealism and weighted reference control
Krea 2 Large Overview
Krea 2 Large is the higher-capacity variant in the Krea 2 family, with lighter post-training and a more textured, flexible output character. It is the stronger overall model when a workflow benefits from higher ceiling, stronger photorealism, more raw visual character, and better handling of motion blur, film grain, low dynamic range, and other less polished looks. It supports text-to-image and image-to-image generation, prompt interpretation strength through the creativity control, and up to 10 weighted reference images with both positive and negative guidance.
Commercial use
How to Use Krea 2 Large
Overview
Krea 2 Large is the higher-capacity variant in the Krea 2 image model family.
It is designed for workflows that want the strongest visual ceiling in the family, especially when rawness, texture, reference fidelity, and photorealistic range matter more than the tighter consistency of the smaller variant.
Strengths
Higher-Ceiling Visual Quality
Krea 2 Large is the more powerful model overall in the family. At its best, it can produce outputs that the Medium variant cannot match, especially in ambitious or visually demanding prompts.
Strong Photorealism
Large is the stronger Krea 2 variant for photorealistic work, especially when the desired result benefits from a less polished and more natural visual character.
Rawer and More Flexible Output Character
Because it uses lighter post-training, Large keeps a more textured, less over-smoothed output profile. This makes it useful for images that benefit from grain, motion blur, low dynamic range, or other more raw photographic qualities.
Weighted Reference Guidance
Krea 2 Large supports up to 10 reference images, with each reference carrying a weight from -2 to +2. This gives finer control than simple reference attachment alone and makes the model useful for both positive steering and deliberate negative guidance away from unwanted looks.
Creativity Control
The exposed API includes a creativity control that adjusts how strongly the model follows prompt interpretation, ranging from raw through low and medium to high. This helps tune whether the result should stay looser and more exploratory or become more tightly aligned to the prompt.
Broad Expressive Range
Large also handles expressive and artistic styles well, so it is not limited to realism. It is useful when one model needs to span photoreal, painterly, editorial, and experimental directions.
Capabilities
Text-to-Image
Krea 2 Large supports prompt-driven image generation across photorealistic, illustrative, cinematic, painterly, and experimental styles.
Image-to-Image and Reference-Guided Workflows
The model also supports image-to-image generation and reference-guided workflows, making it useful for style transfer, guided variation, and source-informed visual direction.
Input and Output
- AIR ID:
krea:krea@2-large - Input: text prompts with optional source image and up to 10 weighted reference images
- Output: generated or reference-guided images
- Creativity control: raw = 0, low = 10, medium = 50, high = 100
- Reference weight range: -2 to +2
- Family role: larger, higher-ceiling Krea 2 variant
Best Fit
- Photorealistic image generation
- Reference-fidelity-heavy workflows
- Raw photographic looks with grain or motion blur
- High-end creative exploration
- Source-informed editorial and campaign work
More models from Krea
Krea 2 Medium is the smaller, faster, and more cost-efficient variant in the Krea 2 family. It applies heavier post-training for especially stable and consistent outputs, making it a strong fit for illustration, anime, painting, graphic design, and other expressive visual workflows. It supports text-to-image and image-to-image generation, prompt interpretation strength through the creativity control, and up to 10 weighted reference images with both positive and negative guidance.
