Skip to gallery
mg_ec999b419bimageen

the character in reference image 1 A female swordsman, wi...

Original prompt

the character in reference image 1 A female swordsman, with cool and beautiful facial features while maintaining facial structure and facial identity on the said character (do not copy the outfit, facial expression and hair) , long black hair, slightly moist hair and messy hair, hair blown by high-speed airflow, light makeup, transparent fair skin, calm and sharp eyes, looking straight into the camera. Wear a loose and slightly wrinkled black shirt with cuffs rolled up at will, a white knitted coat tied around the waist, a bright red white plaid pleated skirt, white over-the-knee stockings, black college leather shoes, and two silver Japanese long knives fastened across your back.

The character holds an unsheathed samurai sword and performs extremely intense swordsmanship movements such as drawing the sword, slashing horizontally, slashing down, blocking, and semi-squatting to accumulate power. The blade is composed of transparent ice crystals, glass and flowing liquid energy. The sharp edge emits a cold purple aura, and strong purple knife gas, arc-shaped sword qi, transparent water flow, ice cracks, glass fragments and high-speed motion trajectories are formed around it. The sword energy surrounds the character's body and passes through the foreground of the picture. Cracks and energy bursting effects appear on the ground, debris splattered, and the picture has an impact of instantly cracking.

Silver gold minimalist studio background, high-profile warm yellow light, front hard flash photography, light gold tone, slight overexposure, skin retains true texture, fine hair, high sharpness, metal and transparent blade with true reflection. Japanese action movie posters, future urban swordsmen, fashion magazine blockbusters and cyber martial arts are visually combined to provide bold compositions, with low camera position, close range and wide angle, with characters centered or tilted, and a strong sense of depth.

The screen is added with black bold English titles and futuristic technical text typesetting. The main titles use high-speed horizontal stretching, scan lines, dislocation tearing, fault drag and ink scratching effects. Part of the text is blocked by the characters and knife air, forming an interspersed relationship between front and back; Add small coordinate lines, barcodes, HUD scales, numbering, technical notes and micro English around. The overall black, white and silver gray color matching is cold, clean, fierce, advanced, with a sense of speed and visual impact, film-level realism, ultra-high definition details, single poster, vertical version 9:16.

English title can be replaced

HEAVEN DIVIDE
SHARP EDGE
GROUND SPLITTER
SHADOW STEP
QUICKDRAW STABS
PHANTOM SLASH
MOONCUT
HELLISH PIERCE

Auxiliary small words:

BLADE MODE
SHADOW PHASE ATTACK
HEAVENLY SLASH
IMPACT LINE
TRIPLE RAPID CUT
ARC SPLIT
FINAL FORM DESCENT
HOLD FRAME

Picture group action change words

The eight pictures maintain the same adult woman, the same outfit and facial features, and each uses different movements: lifting the long knife vertically with both hands, drawing the knife horizontally, squatting with one knee and hitting the ground, crossing the double knives to prepare for attack, and slowly drawing the knife in close range., with the blade horizontally in front of you, swinging the knife at a low angle, kneeling and pressing the long knife with both hands. The characters 'expressions are calm, alert, restrained, sharp and focused. The sword qi forms and English titles are different. Each picture is formed independently, and there is no jigsaw puzzle.

Negative prompt words

Minors, young children's faces, pornographic postures, excessive exposure, see-through clothing, blood, wounds, limb mutilation, real violence scenes, low definition, blurred facial features, plastic skin, deformed fingers, extra arms, extra knives, broken handles, weapon wearing mold, wrong knife holding posture, messy background, warm yellow tones, cartoon feel, low-quality text, garbled code covering the surface, brand logo, watermarks.
Share to ChatGPT
Model:GPT Image
Collected:2026-10-23T12:41:17Z
Categories:portraits, fashion-beauty, characters, fantasy-scifi
Attribution: unknown Prompt text and provenance archived from https://www.meigen.ai . Images and video files remain the property of their authors and are NOT redistributed.

Browse more in the gallery or explore this collection.

Search the archive

Find prompts and collections