These are reusable skeletons, not finished prompts. Copy one, swap the placeholders, and paste it straight into MiniMax Hailuo 2.3, the physics-leading 2026 video model. Every clip runs silent — Hailuo has no native audio, so add sound in your editor afterward.

Two markers, two jobs. The blanks you fill use {curly braces}, like {subject} or {setting} — replace those with your own words and delete any you do not need. The [square brackets] are literal Hailuo Director camera tags the model actually reads, such as [Push in] or [Pan left] — leave those exactly as written. Put up to three camera tags at the very start of the prompt in the order you want them. Keep prompts concise, around 40–60 words, and set aspect ratio, resolution, and duration in the panel. For the full roundup of ready-made prompts, see our best Hailuo prompts collection.

Advertisement

Cinematic shot

These lean into Director camera moves and light to produce a film look from a single text prompt. Lead with the bracket tags, then keep the action to one clear beat.

1. Full-formula cinematic master

[Push in] {subject with 2-3 details}, {one action}, in {setting}. {shot size} shot. {lighting}. {style} style, {mood} mood.

Fill in: {subject} = "a lone astronaut in a scuffed white suit"; {one action} = "turns slowly toward the camera"; {setting} = "a derelict space station corridor"; {shot size} = wide / medium / close-up; {lighting} = "cold blue emergency light"; {style} = cinematic; {mood} = tense. Example filled: [Push in] a lone astronaut in a scuffed white suit turns slowly toward the camera in a derelict space station corridor. Medium shot. Cold blue emergency light. Cinematic style, tense mood. (16:9, 1080p, 6s)

2. Two-move dramatic reveal

[Pedestal up][Push in] {subject}, {one action}, {setting}. {lighting}, {color palette} grade. Cinematic, {mood}.

Fill in: {subject} = "a queen on a stone throne"; {one action} = "raises her eyes to the doorway"; {setting} = "a candlelit great hall"; {color palette} = "warm amber and deep shadow"; {mood} = regal. The two tags run in order: lift, then close the distance. (16:9, 1080p, 6s)

3. Static portrait with emotion

[Static shot] Close-up of {person with 2-3 details}, {one facial action}, looking {direction}. {lighting}. {emotion} mood.

Fill in: {person} = "an older man with a grey beard and wool cap"; {one facial action} = "a slow, warm smile forming"; {direction} = "off-camera left"; {lighting} = "soft window light"; {emotion} = "quiet, reflective". A held frame keeps the face stable. (16:9, 1080p, 6s)

Product & ad

These frame a single product cleanly with controlled motion. Lead with one gentle camera tag and let the surface and light do the selling. For deeper technique, see how to prompt Hailuo for realistic video.

4. Studio product turntable

[Static shot] {product} centered on {surface}, slowly rotating. Clean {background color} studio background, soft {lighting} with gentle reflections. Premium, minimal.

Fill in: {product} = "a matte-black wireless headphone"; {surface} = "a polished stone slab"; {background color} = "off-white"; {lighting} = "soft key from upper left". The held camera lets the spin read cleanly. (1:1, 1080p, 6s)

5. Product hero push-in

[Push in] {product} on {surface}, {subtle prop motion} around it. {lighting}, {color accent} rim light. Sleek, aspirational.

Fill in: {product} = "a glass perfume bottle"; {subtle prop motion} = "petals drifting down"; {lighting} = "dark moody key light"; {color accent} = "gold". The push builds toward the label. (16:9, 1080p, 6s)

6. Product in-use lifestyle

[Tracking shot] Close-up of {hands or person} using {product} to {action} in {setting}. Shallow focus on the product. {lighting}, natural {mood}.

Fill in: {hands or person} = "a woman's hands"; {product} = "a ceramic pour-over kettle"; {action} = "pour water in a slow spiral"; {setting} = "a sunlit kitchen"; {mood} = calm. Tracking reads authentic for lifestyle. (9:16, 1080p, 6s)

B-roll

Cutaways and texture inserts. Keep the motion to a single element so Hailuo's physics stays crisp across the frame.

7. Establishing wide drone shot

[Pedestal up] Extreme wide establishing shot of {place}, {one element moving}, {time of day}. {lighting}, {weather}. Cinematic, {mood}.

Fill in: {place} = "a fog-wrapped mountain village"; {one element moving} = "smoke rising from chimneys"; {time of day} = golden hour; {weather} = light mist; {mood} = serene. The rise mimics a drone ascent. (16:9, 768p, 10s)

8. Tracking b-roll follow

[Tracking shot] Medium shot following {subject} as they {action} through {setting}. Shallow depth of field. {lighting}, {color palette} grade. {mood}.

Fill in: {subject} = "a courier on a bicycle"; {action} = "weaves between market stalls"; {setting} = "a crowded old-town lane"; {color palette} = "warm amber and teal"; {mood} = lively. Great for movement-driven cutaways. (16:9, 1080p, 6s)

9. Macro detail insert

[Zoom in] Extreme close-up macro of {subject detail}, {subtle action}. {lighting}, soft focus falloff. {mood} mood.

Fill in: {subject detail} = "coffee crema swirling in a cup"; {subtle action} = "steam curling upward"; {lighting} = "warm side light"; {mood} = cozy. Perfect for texture-rich inserts. (16:9, 1080p, 6s)

Advertisement

Image-to-video motion

With Image-to-Video the uploaded still already supplies subject, setting, and lighting, and the clip follows the source aspect ratio. These skeletons describe motion and camera only — nothing else.

10. Gentle camera on a still

[Push in] {one subject} {subtle motion}. Everything else stays still.

Fill in: {one subject} = "the woman's hair"; {subtle motion} = "drifts in the breeze". Swap the tag for [Pan left] or [Tilt up] to match your framing. (matches source, 1080p, 6s)

11. Subject enters motion

[Tracking shot] {subject} begins to {action}. Keep the background static.

Fill in: {subject} = "the car"; {action} = "drive forward slowly". Motion-only keeps the source scene intact while the tracking tag follows it. (matches source, 1080p, 6s)

12. Living-still micro loop

[Static shot] Only {area} moves: {subtle looping motion}. Everything else frozen.

Fill in: {area} = "the waterfall"; {subtle looping motion} = "water flowing downward". A held camera plus one moving region reads like a clean cinemagraph. (matches source, 1080p, 6s)

13. Slow atmospheric drift

[Truck right] Camera glides across the scene while {ambient element} {subtle motion}. Subject stays in place.

Fill in: {ambient element} = "dust motes in the light"; {subtle motion} = "float gently". The lateral truck adds depth without altering the subject. (matches source, 768p, 10s)

Action & physics

Hailuo 2.3 leads on physical realism, so these push weight, momentum, and collision. Keep it to one action so the simulation stays believable.

14. Weighted action beat

[Tracking shot] Medium shot of {character} {one physical action} in {setting}, wearing {wardrobe}. {lighting}, {mood}.

Fill in: {character} = "a young dancer"; {one physical action} = "spins once and holds a pose"; {setting} = "an empty warehouse"; {wardrobe} = "a flowing red dress"; {mood} = dramatic. One action keeps limbs stable. (16:9, 1080p, 6s)

15. Impact & collision

[Static shot] {object} {collides or falls} onto {surface}, {resulting physics}. {lighting}, high detail. {mood}.

Fill in: {object} = "a ripe watermelon"; {collides or falls} = "drops and bursts"; {surface} = "a concrete floor"; {resulting physics} = "shards and juice scatter outward"; {mood} = visceral. Hailuo resolves the splatter cleanly on a held frame. (16:9, 1080p, 6s)

16. Motion through a crowd

[Tracking shot][Shake] {subject} {runs or moves} through {dense setting} as {crowd reacts}. {lighting}, {mood}.

Fill in: {subject} = "a sprinter"; {runs or moves} = "bursts forward"; {dense setting} = "a packed train platform"; {crowd reacts} = "commuters turn to look"; {mood} = urgent. The shake adds handheld energy to the chase. (16:9, 1080p, 6s)

Vertical social 9:16

Vertical clips need the subject centered and one punchy action in the first second. Keep camera tags short. For a fast reference on tags and framing, see the Hailuo prompt cheat sheet or our cinematic b-roll prompts.

17. Vertical hook shot

[Push in] Vertical framing. {subject} centered, {bold action} immediately. {lighting}, {high-energy mood}.

Fill in: {subject} = "a chef"; {bold action} = "flips a pan of flaming vegetables"; {lighting} = "warm kitchen glow"; {high-energy mood} = "energetic". Front-load the action to stop the scroll. (9:16, 1080p, 6s)

18. Vertical talking-head b-roll

[Static shot] Vertical medium close-up of {person} {subtle action} in {setting}. {lighting}, {mood}.

Fill in: {person} = "a young creator"; {subtle action} = "nodding and gesturing lightly"; {setting} = "a plant-filled room"; {mood} = friendly. The clip is silent — add voiceover in your editor. (9:16, 1080p, 6s)

19. Vertical POV walk-through

[Push in] Vertical POV moving forward through {environment}, {one element reacting}. Steady pace. {lighting}, {mood}.

Fill in: {environment} = "a neon-lit night market"; {one element reacting} = "steam rising from food stalls"; {lighting} = "colorful neon reflections"; {mood} = immersive. POV reads well on mobile. (9:16, 768p, 10s)

20. Vertical product flatlay

[Tilt down] Vertical top-down of {product} on {surface}, {props} around it, {subtle motion}. {lighting}, clean and bright.

Fill in: {product} = "a skincare jar"; {surface} = "marble"; {props} = "eucalyptus sprigs"; {subtle motion} = "a hand placing the jar down". Ideal for social product teasers. (9:16, 1080p, 6s)

Frequently Asked Questions

How do I use these Hailuo templates?

Copy a template, replace every {curly-brace} blank with your own words, delete any blank you do not need, and keep the literal [Camera move] tags as written. Paste the finished sentence into Hailuo 2.3. Set aspect ratio, resolution, and duration in the interface, not in the text.

Why do templates use {curly braces} and [square brackets] differently?

Square brackets are literal Hailuo Director syntax the model reads, such as [Push in] or [Pan left]. To avoid confusion, the fill-in blanks in these templates use {curly braces} instead. Replace the {curly} parts with your words and leave the [square] camera tags exactly as written.

Do the camera tags go first in a Hailuo prompt?

Yes. Place Director camera moves in square brackets at the very start of the prompt, up to three in the order you want them performed, for example [Push in][Pan right]. Then describe subject, action, setting, and lighting.

What duration and resolution work best in Hailuo 2.3?

Choose 6s or 10s and 768p or 1080p. Note that 1080p is capped at 6s, so pick 768p if you need the full 10s length. Set these in the panel, never in the prompt text.

What is the difference between Text-to-Video and Image-to-Video here?

Text-to-Video builds the whole scene from your words and defaults to 16:9. Image-to-Video starts from an uploaded still that already supplies subject, setting, and lighting, so those prompts describe only motion and camera and follow the source aspect ratio.

Can I add audio in Hailuo 2.3?

No. Hailuo 2.3 has no native audio generation, so every clip is silent. Add voiceover, music, or sound effects afterward in your editor. Writing sound cues into the prompt has no effect.

Advertisement