MiniMax H3 / Hailuo 3.0 · T2VA

One Stage Performance Across Two Models

The same blonde performer, arena stage, and complete direction are used for a side-by-side model comparison. Native-video output: 27s · 2160×2880. The creator published the complete prompt, preserved here verbatim.

Generation mode
T2VA
Model
MiniMax H3 (creator-labeled)
Output
27s · 2160×2880 · portrait
Tags
Model Comparison · Comparative
Prompt provenance
Creator, verbatim

Original Prompt

Verbatim prompt as published by the source; not translated, rewritten, or completed.

— A young Caucasian woman, 177 cm tall, with long, straight, silky light-blonde hair and neat, airy bangs; she has light gray, clear eyes and delicate, cool-looking features. She is wearing a white, fitted, short, pleated V-neck spaghetti-strap tank top, blue fitted jeans, a brown leather belt with a large vintage-style silver buckle, and dark brown leather knee-high boots.
 — white wireless microphone.
 — round stage in the center of the arena, surrounded by crowd on all sides, LED screen above.
 — music track, exactly this track must play throughout the entire scene.

Mira performing on stage to the track — not singing, moving like a rock star, dynamic chaotic movements, aggressively moving to the beat, sharply headbanging. White microphone in hand. Not looking at the camera. Shooting only mira — super close-up, cuts allowed. No zooms — camera always physically right next to mira on stage, heavily tilted, handheld. Dark cinematic lighting — top light, soft warm light barely touching her face, deep shadows below. Realistic skin. ARRI Alexa 35, anamorphic lenses. No 3D, no VFX.

View original source · Original X post · @ou_zhen599