Krea just shipped Generative Sliders for its Krea 2 image model , a new set of controls that let you tune the intensity, complexity, and movement of any image you generate, directly from the image tool. No prompt rewriting, no guessing. Just drag a slider and watch the output shift.

The problem with prompting for feel

Text prompts are great at describing what is in an image. They are notoriously bad at describing how much. Telling a model to make something "more dynamic" or "slightly more complex" is a coin flip , you might get what you want, or you might get a completely different image that happens to match the word. Most AI image models optimize for one thing: doing exactly what you tell them. You describe a scene, they render it. The more precise your prompt, the more literal the output.

Generative Sliders sidestep that entirely. Instead of encoding intent into language, you move a slider. The three controls map to distinct visual dimensions:

• Intensity , how visually striking or bold the output feels
• Complexity , the density of detail, elements, and visual information in the scene
• Movement , the sense of motion or dynamism in the image

This fits squarely into Krea's broader philosophy around Krea 2. Krea 2 was trained completely from scratch to give users full control over the look, feel, and creative direction of every image , with particular strength in artistic and expressive directions other models tend to flatten. Generative Sliders extend that control into dimensions that were previously only accessible through trial-and-error prompting.