Image to Video Prompt Examples: 40 Prompts That Work

A good image-to-video prompt describes only what should move: the subject's action, the camera move and the light change. It does not re-describe the photo. Below are 40 copy-paste prompts in five groups (portraits, products, landscapes, pets, couples), each tagged with the model it suits. On VdoBloom's Image to Video tool a clip costs from 15 credits (Hailuo 2.3 Standard, 6 seconds, 768p), 25 credits for Veo 3.1 Fast (8 seconds with sound) and 40 for Kling 3.0 Omni (5 seconds, 720p). New accounts get 10 free credits, there is no subscription, and one-time pack credits never expire.

Updated October 31, 2026.

Quick facts

• Prompt length: 1 to 3 sentences. Longer prompts make models redraw the photo.
• Formula: subject action + camera move + one atmosphere detail + a "keep" clause.
• Cheapest test: Hailuo 2.3 Standard, 15 credits for 6 seconds at 768p.
• Sound in one pass: Veo 3.1 Fast, 25 credits for 8 seconds with audio; Veo 3.1 High Quality 100 credits.
• Best motion control: Kling 3.0 Omni, 40 credits for 5 seconds at 720p.
• Failed generations are refunded automatically.

Which model should I use for image to video?

Model	Suits	Credits (example)
Hailuo 2.3 Standard	Quick tests, gentle portrait and landscape motion	15 (6s, 768p)
Kling 3.0 Omni	Precise body motion, products, camera orbits	40 (5s, 720p)
Veo 3.1 Fast	Clips that need sound: speech, ambience, pets barking	25 (8s)
Seedance 2	Expressive people, couples, multi-beat action	Shown before you generate
Wan 3.0	Stylised and scenic shots	Shown before you generate

How do I write an image-to-video prompt?

1. Upload a sharp photo with the subject clearly separated from the background.
2. Write one action verb for the subject ("turns", "pours", "sways").
3. Add one camera move ("slow push-in", "gentle orbit left", "static tripod").
4. Add one atmosphere detail (wind, steam, light flicker).
5. End with a keep clause: "keep the face, outfit and background unchanged."
6. Render once on a cheap model, then re-run the winner on a higher-quality one.

Portrait prompts (1 to 8)

1. She slowly turns her head toward the camera and smiles softly, hair moving in a light breeze, slow push-in, keep her face and outfit unchanged. Suits: Hailuo 2.3.
2. He blinks, takes a calm breath and looks slightly off-camera, static tripod shot, soft window light, natural skin texture. Suits: Kling 3.0 Omni.
3. The woman looks into the lens and says, "Okay, let's start," then laughs quietly, room tone in the background. Suits: Veo 3.1 Fast.
4. Golden-hour sun flickers across his face as leaves sway behind him, camera drifts slowly to the right, keep the background unchanged. Suits: Hailuo 2.3.
5. The graduate tosses her cap upward and catches it, cheering, slight handheld camera, confetti drifts past. Suits: Seedance 2.
6. Old photo comes to life: the man nods gently and his eyes crinkle into a smile, very subtle motion, film grain stays. Suits: Kling 3.0 Omni.
7. Professional headshot: she adjusts her blazer, settles and gives a confident nod, static camera, studio background unchanged. Suits: Hailuo 2.3.
8. Rain begins to fall around him, droplets catch the streetlight, he pulls up his hood, slow dolly back. Suits: Wan 3.0.

Product prompts (9 to 16)

1. The bottle rotates slowly on its base, a soft highlight sweeps across the glass, black studio background, label stays sharp and readable. Suits: Kling 3.0 Omni.
2. Steam rises from the coffee cup, camera pushes in slowly toward the rim, warm morning light, keep the mug design unchanged. Suits: Hailuo 2.3.
3. Water droplets run down the cold can, a slow orbit around it, condensation glistens, sound of a can opening at the end. Suits: Veo 3.1 Fast.
4. The sneaker lifts off the ground and spins once in mid-air, then lands softly, dust puffs out, static camera. Suits: Kling 3.0 Omni.
5. Cream swirls slowly in the open jar, camera tilts down from the lid to the texture, clean white surface. Suits: Seedance 2.
6. The watch hands tick forward, light glints off the crystal, slow macro push-in, dark velvet background. Suits: Kling 3.0 Omni.
7. Petals fall around the perfume bottle as a gentle wind passes through, slow orbit right, keep the bottle shape exact. Suits: Wan 3.0.
8. A hand reaches in and picks up the phone from the desk, screen lights up, static overhead shot. Suits: Seedance 2.

For full ad structures, see food and product ad video prompts.

Landscape and real-estate prompts (17 to 24)

1. Clouds roll over the mountain ridge, mist drifts through the valley, slow aerial push forward. Suits: Hailuo 2.3.
2. Waves wash gently onto the beach and pull back, palm leaves sway, sound of surf and distant gulls. Suits: Veo 3.1 Fast.
3. Living room walkthrough: camera glides slowly from the doorway toward the window, curtains move slightly, keep furniture unchanged. Suits: Kling 3.0 Omni.
4. Sunset turns to blue hour over the city skyline, windows light up one by one, static wide shot. Suits: Wan 3.0.
5. Snow starts falling over the cabin, smoke curls from the chimney, slow push-in toward the lit window. Suits: Hailuo 2.3.
6. The waterfall pours steadily, spray drifts across the frame, camera tilts up from the pool to the top. Suits: Kling 3.0 Omni.
7. Pool water ripples in the breeze, sun reflections dance on the villa wall, slow lateral slide left. Suits: Hailuo 2.3.
8. Northern lights ripple across the night sky above the lake, stars stay still, very slow push-in. Suits: Wan 3.0.

Pet prompts (25 to 32)

1. The dog tilts its head, ears perk up, then it wags its tail happily, static camera at its eye level. Suits: Hailuo 2.3.
2. The cat slowly blinks at the camera, then yawns and stretches its paws forward. Suits: Kling 3.0 Omni.
3. The puppy barks twice at the camera and bounces on its front paws, garden sounds in the background. Suits: Veo 3.1 Fast.
4. The cat bats at a falling leaf and misses, then looks around as if nothing happened. Suits: Seedance 2.
5. The dog runs toward the camera across the grass, ears flapping, slow-motion feel, camera tracks backward. Suits: Kling 3.0 Omni.
6. The parrot fluffs its feathers and turns its head side to side, soft morning light. Suits: Hailuo 2.3.
7. The rabbit's nose twitches and it hops once toward the lens, static close-up. Suits: Hailuo 2.3.
8. The dog wearing sunglasses slowly lowers its head and looks over the frames at the camera, comic timing. Suits: Seedance 2.

More in pet and dog video prompts and cat video prompts.

Couple and wedding prompts (33 to 40)

1. The couple turns toward each other and touch foreheads, smiling, slow orbit around them, keep both faces unchanged. Suits: Kling 3.0 Omni.
2. Bride and groom walk hand in hand toward the camera, her veil lifts in the wind, camera tracks backward. Suits: Seedance 2.
3. They clink glasses and laugh, string lights glow behind them, sound of a quiet party. Suits: Veo 3.1 Fast.
4. He spins her once under his arm, her dress flares, warm sunset backlight, static camera. Suits: Seedance 2.
5. The couple sits on the bench and she rests her head on his shoulder, autumn leaves fall, slow push-in. Suits: Hailuo 2.3.
6. Old wedding photo comes alive: they glance at each other and smile, subtle motion, keep the sepia tone. Suits: Kling 3.0 Omni.
7. They share an umbrella in light rain and look up at the sky, puddles reflect city lights. Suits: Wan 3.0.
8. Anniversary moment: he says, "Happy anniversary," and she covers her mouth in surprise. Suits: Veo 3.1 Fast.

If the photo has two people, describe each person's action separately ("she laughs, he pulls her closer"). One shared verb is the most common reason one face drifts.

Why does my image-to-video clip ignore the prompt?

• Too many actions. Keep one main action per clip.
• You described the photo. The model already sees it; describe motion only.
• The move fights the frame. A tight crop cannot "pull back to reveal the room".
• Wrong model. Need speech or sound? Use Veo 3.1. Need exact motion? Use Kling.

See AI video not following the prompt and camera movement prompts for deeper fixes.

FAQ

What is a good prompt for image to video?

One subject action, one camera move, one atmosphere detail and a keep clause, in one to three sentences. Example: "She turns toward the camera and smiles, hair moving in the breeze, slow push-in, keep her face unchanged."

Which AI model is best for image to video?

It depends on the shot. Kling 3.0 Omni gives the most controlled motion, Veo 3.1 adds sound and speech, Hailuo 2.3 is the cheapest way to test, and Seedance 2 handles expressive people well.

How much does image to video cost on VdoBloom?

From 15 credits (Hailuo 2.3 Standard, 6 seconds at 768p). Veo 3.1 Fast is 25 credits for 8 seconds with sound and Kling 3.0 Omni is 40 credits for 5 seconds at 720p. New accounts get 10 free credits.

Can I add sound to an image-to-video clip?

Yes. Pick Veo 3.1 and describe the sound or put spoken words in quotes.

Do I need a subscription?

No. Buy a one-time credit pack whenever you need it; pack credits never expire.