Couple sitting closely on park bench at dusk - AI generated image prompt sample

Couple sitting closely on park bench at dusk - AI Image Prompt

Generates a realistic couple sitting closely on park bench at dusk image with natural body language, soft environmental lighting, and human-accurate proportions. Works best in Gemini, ChatGPT, Grok, Flux, Z-Image and other models that prefer natural language prompts.

Prompt

A romantic portrait of a couple sitting side by side on a park bench in the evening. 

Subject: A man and a woman with dark hair; the man has short cropped hair, clean-shaven face, and neutral expression; the woman wears her brown hair loose over shoulders, soft smile gazing at him. 

Clothing: The man is dressed entirely in black a tailored blazer, plain round-neck t-shirt, slim-fit trousers sophisticated monochrome look; the woman wears an elegant sleeveless wrap-style dress with a V-neckline and gathered waistband. 

Action: They are seated close together on a wooden park bench; their hands rest gently clasped between them in front of her lap; both gaze tenderly into each other's eyes, conveying intimacy and quiet connection. 

Environment: Soft-focus background reveals lush green trees and blurred streetlights glowing faintly behind foliage suggesting an urban park at night or dusk with bokeh effect enhancing depth and romance. 

Lighting: Gentle ambient evening light illuminates their faces softly from above; subtle shadows define facial contours without harshness, creating a warm, intimate mood balanced by cool-toned background glow. 

Camera: Medium shot framing both subjects clearly against blurred backdrop; shallow depth of field keeps focus tightly on them while smoothing surroundings into dreamy bokeh circles (lens flare hints).

Category

Tags

Medium ShotFull BodyBrown HairMaleFemaleCasual WearPortrait ShotOutdoorsRealistic

How to use

Generate a new image

  1. Copy the prompt and paste it into your preferred image model.
  2. Adjust details like subject, lighting, or style if needed.
  3. Generate and iterate to refine the result.
  4. Works best with models that support detailed natural language prompts, such as Gemini, Grok, ChatGPT, Flux, or Z-Image.

Edit your uploaded image

  1. Upload your image in a model like Gemini, Grok or ChatGPT.
  2. Paste the edit prompt into the text box and submit.
  3. Review the result, most edit prompts work out of the box.
  4. If needed, tweak the prompt and repeat until you get the desired output.
  5. Note: In our testing, Gemini and Grok tend to give better results for edits.

Related Prompts

More prompts with similar categories and tags.