Original document
GPT Image 2 · en
A cinematic, hyper-realistic equestrian portrait capturing the golden moment, featuring a horse {argument name="horse color" default="Dark chestnut"} Ma and one{argument name="rider gender" default="young men"} They stood closely together in the quiet outdoor paddock at sunset. The horses were tall and elegant, shown in a three-quarter profile with a close-up of their chests and above, their coats glossy, their long, dark manes flowing, their ears alert and erect, and they wore black leather reins and harnesses; a portion of an English saddle was visible on the left side of the image. The riders wore well-fitting {argument name="rider outfit" default="Black long-sleeved riding shirt and black riding breeches"} Standing to the right of the horse, leaning forward, one hand gently cradling the horse's cheek and muzzle, the pose is intimate. The composition employs a close-up portrait, with the horse occupying the left half of the frame and the rider on the right, their faces almost touching, emphasizing emotional connection and trust. The scene is set in a fenced meadow, with a blurred wooden fence, distant rolling hills, and an orange sky illuminated by the low-hanging sun on the horizon. Warm backlighting, rich amber highlights, realistic skin and fur textures, shallow depth of field, a high-end advertising photography style, natural shadows, high contrast, and a sophisticated, luxurious editing feel are used. The overall atmosphere should be warm, tranquil, and refined, like an advertisement for a high-end lifestyle of equestrian fashion or horse care.How to use
- Copy the complete original prompt, including any parameters.
- Replace the subject, text and composition with your own requirements.
- Use a compatible external tool; check its current access conditions and review the result.
Before you start
- Check whether a reference image is required. A compatible model label does not mean this example was generated with that model.
- Keep names, figures and visible text accurate; results can vary between model versions.
