Copy and adapt practical MiniMax H3 prompts for text-to-video, image animation, camera movement, product ads and character-focused cinematic clips.
Quick answer
At a glance
Copy and adapt
Use each prompt as a working shot brief. Replace the subject and constraints with details from your own scene.
Text to Video
A lone street photographer pauses beneath a red awning in a rain-soaked night market. Slow push-in from a medium-wide frame, neon reflections ripple across the pavement, shallow depth of field, restrained crowd motion, quiet city ambience, cinematic contrast.
Text to Video
A matte-black fragrance bottle stands on dark stone. Begin with an extreme macro of condensation, then make one slow orbit into a centered hero frame. Warm rim light, precise reflections, soft studio atmosphere, no text deformation.
Image to Video
Preserve the uploaded portrait composition and facial identity. Add a subtle blink, natural breathing and gentle hair movement from a light breeze. Slow camera drift to the right, keep the background locked and maintain the original color grade.
Image to Video
Keep the product shape, label and camera angle unchanged. Shift the highlight gradually across the surface, add a controlled push-in and a soft shadow movement. End on a two-second clean hero hold.
First & Last Frame
Connect the closed package in the first frame to the open package in the last frame with one smooth dolly forward. The lid opens naturally as warm afternoon light moves across the table. Preserve product position and avoid cuts.
First & Last Frame
Transform the daylight city frame into the supplied night frame in one continuous locked-off shot. Windows illuminate gradually, traffic trails appear, sky color changes smoothly, no camera jump or scene replacement.
Subject Reference
Use the uploaded face as the same lead character walking through a busy open-air market. Preserve facial features, hair and jacket. Steady tracking camera, natural crowd movement, soft late-afternoon light, one continuous shot.
Subject Reference
Keep the referenced subject recognizable while the camera performs a gentle half orbit from front three-quarter to profile. The subject remains still, wardrobe and proportions stay consistent, neutral studio light, no identity change.
01
Start from the input you already trust. Text is fastest for open concepts; an image locks the look; two frames lock the transition endpoints; a subject reference protects identity.
02
Write one compact shot direction: [subject + action] in [setting]. [primary camera move], [lighting and mood]. Add mode-specific instructions for what must stay stable.
03
Concrete camera terms are more useful than a stack of style adjectives. Pick a static wide, push-in, pan, tracking move or orbit and let that move serve the action.
04
Replace the subject, setting and constraints with your own brief. If you upload an image, remove prompt details that fight the visible composition.
Common questions
Include the subject, action, setting, one camera move, lighting and mood. Add a clear consistency instruction for image, frame-pair or subject-reference workflows.
Yes, but results improve when you replace the subject and constraints with details from your own scene and reference assets.
Long enough to define one coherent shot. Several concrete sentences normally work better than a long list of disconnected adjectives.
Choose Subject Reference, upload one clear face image and separate identity instructions from wardrobe, action and environment.
Describe the motion connecting the supplied endpoints. Do not ask the model to invent a third unrelated composition.
Yes. Use terms such as static wide, push-in, pan, tracking shot or orbit, preferably one primary move per clip.
Related resources
Start with text or an image, direct one clear shot, then preview and refine the result in your browser.