Up to 40% less per creditEligible accounts: 30 signup credits
  • Home
  • Agent
  • Images
  • Videos
  • Prompts
  • Assets
  • Plugins

Loading…

FilmLune

From real examples to AI image and video creation.

Models

Image models

  • GPT Image 2
  • GPT Image 2.5
  • Midjourney V8.1
  • Midjourney V8.2
  • Nano Banana Pro
  • FLUX 3 Image
  • Qwen Image 3.0 Pro
  • Seedream 5.0 Lite

Video models

  • Grok Video 1.5
  • MiniMax H3
  • Seedance 2.0
  • Seedance 2.0 Mini
  • Seedance 2.5
  • Dola AI alternative
  • Kling 4.0
  • Wan 3.0
  • Gemini Omni Flash 1.1
  • HappyHorse 1.1
  • FLUX 3 Video
  • MiniMax H3 Max

Create

  • Open Studio

Tools & Skills

  • Features
  • MCP server

Resources

  • All resources
  • Changelog

Prompts

  • Explore prompts
  • Image prompts
  • Video prompts

FilmLune

  • Contact
  • Pricing

Community

  • Follow FilmLune on X ↗
  • Join Discord ↗

Legal and privacy

  • Terms
  • Refunds
  • Privacy Policy
  • GDPR
  • CCPA
  • Cookie choices
  • Sharing rules

© FilmLuneKeep creative work and trustworthy information close at hand.

Home

Loading…

Prompts/AI video prompts/A giant dog towers over a city street
1 / 2
esc
1 / 2
FilmLune
Grok Video 1.5

PROMPT

Use case: photorealistic-natural. Create ONE starting frame for a short image-to-video film. Comic disaster-film mood. Portrait 4:5 composition.

A handheld pedestrian's-eye view down a deep Manhattan-like avenue lined closely on both sides by tall office buildings, in hazy overcast daylight. At the far end of the street, one colossal living dog towers several stories above the people and taxis, centered between the buildings and facing the camera. The dog has short sandy-tan fur, floppy ears, a broad dark charcoal muzzle and a black nose, with a gentle, slightly worried face; it resembles a giant real shepherd-mastiff, not a monster. Its huge body and front legs are visible behind the street traffic, leaving room for it to take a slow step toward the camera later.

A few pedestrians run toward the viewer in the near foreground, dwarfed by the dog; several ordinary yellow taxis fill the lower left and street. Convey the vast change in scale at a glance. Candid smartphone-video realism, slightly soft and kinetic, natural urban daylight. One dog only. Do not add helicopters, captions, logos, text, gore, damage, explosions, or collage panels.
Create a 10-second photorealistic disaster-comedy video with three short shots. Use the supplied image as the opening frame. Keep the same one gigantic, gentle-looking sandy-tan short-haired dog with floppy ears and a dark muzzle in every shot. The dog is several stories tall, enormous next to people, yellow taxis, and buildings. The dog is not attacking anyone. Handheld, candid, slightly shaky smartphone footage, realistic natural daylight.

0–3.7 seconds: Street level, facing down the avenue. People in the foreground run toward the camera as the colossal dog slowly walks toward them between the buildings. Show one massive front paw lifting and landing on the street; taxis remain tiny beside it. The camera retreats shakily and the dog's big face gets closer, looking calm and slightly puzzled.

3.7–6.8 seconds: Hard cut to a view through a high-rise office window. The same dog's giant head rises outside near the window, almost level with the upper floors. It turns its head and looks curiously toward the window. The city skyline is behind it and a tiny helicopter crosses high in the distance.

6.8–10 seconds: Hard cut back to a very low street-level angle looking up between skyscrapers. The same dog's immense head and chest fill the center while it gently raises its muzzle and looks up at two little helicopters circling overhead. Preserve the dog's recognizable face and enormous scale through all three shots.

Natural urban crowd and traffic ambience with distant helicopter rotor sounds; no narration. No captions, logos, extra dogs, monster teeth, destruction, or attack. Do not change the animal's color, facial markings or size.

In this collection

▶

Prompt guide · AI video

A giant dog towers over a city street — AI Video Prompt

A colossal tan dog walks between city buildings, peers through an office window and looks up at helicopters in three connected shots.

  • ModelGrok Video 1.5
  • Use caseStoryboards, video & motion
  • StyleHandheld giant-animal comedy
  • CollectionDisaster comedy videos

Keeping one enormous dog recognizable across three shots

At street level, pedestrians run toward the camera while a large tan dog moves between tall buildings and yellow taxis. A hard cut moves inside an office, where the dog's face appears outside the window and turns sideways. Another hard cut returns to a low street view as the dog lifts its muzzle toward small helicopters overhead. The dog remains gentle-looking; the comedy comes from its scale and the surrounding reactions. The opening image establishes the dog's tan coat, floppy ears and dark muzzle. The video prompt repeats those identifying features across three timed views. Each viewpoint adds a separate scale cue: people and taxis at street level, upper-floor windows beside the head, and helicopters above the muzzle. Explicit cut instructions let one generation change perspective without adding an unseen reference image for each shot.

How to adapt the method

Replace the animal or city architecture in the image prompt and regenerate the starting frame. Repeat the same coat, face and scale instructions across all three video beats. Keep each shot focused on one readable action and inspect continuity at both cuts. For the supported 2:3 reuse option, leave more vertical space and reposition the foreground traffic; do not stretch or crop the supplied 848 × 1072 clip to imply it was generated in that format. Adapted public recipe; the supplied result is not an exact reproduction of these public prompts.

How to use this prompt

  1. 1

    This is an image-to-video recipe. Create the opening image first, then use that output as the input to the video step. Keep the image prompt and motion prompt separate.

  2. 2

    Open the selected recipe in Studio with its model and starting settings. Check the available version, frame and resolution before generating; a size written in the source prompt is not an output setting or a guarantee.

  3. 3

    Use the customization fields below to rewrite the brief: Opening frame / Delivered dimensions / Delivered duration. Keep the composition and other constraints that make this example useful. Check the current credit quote before submitting.

What to check in your result

Watch the entire clip. Compare the main action, camera movement and scene continuity with the brief, and check for changing faces, distorted hands or unexpected cuts. A similar prompt can give a different result; refine the motion rather than assuming an exact copy.

Review the selected prompt and settings in Studio, then submit when you are ready.

Variables to customize

Opening frame
One giant tan dog, pedestrians and yellow taxis in a city avenue
Delivered dimensions
848 × 1072
Delivered duration
10.041667s

Source and attribution

This example is attributed to FilmLune.

View the source on the public source

Report a rights or attribution issue

More recipes to explore

View all AI video examples
A duck float at a rooftop pool edgeA duck float at a rooftop pool edgeSeven Wonders Selfie Travel HyperlapseSeven Wonders Selfie Travel HyperlapseFPV Dive Through a Jungle Temple to a Golden Underground SunFPV Dive Through a Jungle Temple to a Golden Underground SunTiger-Hat Kitten and Yellow Balloon: A 30-Second Prompt AdaptationTiger-Hat Kitten and Yellow Balloon: A 30-Second Prompt Adaptation