-Click%20to%20View-lightgrey){kind=icon}
-Click%20to%20View-lightgrey){kind=icon}
A curated collection of high-quality video generation prompts for Google's Gemini Omni
⚠️ Copyright Notice: All prompts are collected from the community for educational purposes. If you believe any content infringes on your rights, please open an issue and we will remove it promptly.
- 🤔 What is Gemini Omni?
- 📊 Statistics
- ⭐ Featured Prompts
- 🎬 All Prompts
- 🤝 How to Contribute
- 📄 License
- 🙏 Acknowledgements
- ⭐ Star History
Gemini Omni is a video generation model developed by Google. It is the industry's first model supporting simultaneous quad-modal input — image, video, audio, and text.
Key Features:
- 🎥 Text-to-Video — Generate videos from text descriptions
- 🖼️ Image-to-Video — Animate static images into dynamic videos
- 📹 Video-to-Video — Transform and extend existing videos
- 🎵 Audio-Driven — Generate videos driven by audio input
- 📐 Up to 1080p resolution, 4–15 seconds duration
- 🔊 Auto dubbing & scoring — Automatic voiceover and background music
| Metric | Count |
|---|---|
| 📝 Total Prompts | 25 |
| ⭐ Featured Prompts | 0 |
| 🔄 Last Updated | 2026-06-01 |
📝 Sorted by publish date (newest first)
A creative prompt using descriptive keywords and emojis to generate a stable, cinematic tarot card-inspired video featuring a mythical Rusalka theme.
Reflection-Rusalka «🌳🧜♀️🪞»
Author: Elis Satu | Source: Link | Published: May 28, 2026
A creative template for Gemini Omni that recreates iconic movie and TV scenes within the stylized aesthetic of The Sims videogame, including unique graphics and dialogue.
[Movie/TV Show] iconic scene, but it's a Sims videogame
[scene description]
Sims videogame graphics, and Sims silly dialogue
Author: TechHalla | Source: Link | Published: May 28, 2026
A master prompt for Gemini Omni to create a hyper-lapse travel video that swaps an uploaded character into 20 distinct environments while keeping identity consistent.
Create a 10s hyper-lapse selfie-travel video of uploaded character. Strict identity consistency across all locations and professions. 20 distinct absurd professional roles and extreme environments worldwide with hard
Author: ZeFred.AI | Source: Link | Published: May 28, 2026
A cinematic prompt for generating a high-detail weather simulation from a satellite perspective, featuring various extreme climate systems with scientific labels and realistic sound design.
Create a 10-second cinematic weather simulation visualized from satellite perspective showing Earth’s most extreme climate systems: hurricanes, monsoons, lightning storms, auroras, tornado outbreaks, sandstorms, volcanic ash clouds. Scientific labels appear in clean white Noto Sans Google Font like a live meteorological interface. High-detail cloud simulations and atmospheric realism. Sound design layers deep thunder, wind turbulence, distant emergency radio chatter, atmospheric drones, and cinematic tension music. No dialogue.
Author: Heather Cooper | Source: Link | Published: May 28, 2026
A prompt to generate a 10-second high-end travel poster video featuring a panoramic moving strip and stylized typography for a specific city.
[CITY] = KTOKYO
Create a 10-second premium travel-poster style video for [CITY].
Use a soft ivory poster background, a thin moving panoramic strip at the top, and a huge bold centered sans-serif title: “[CITY]”.
Author: マグマグ | Source: Link | Published: May 28, 2026
A simple prompt for generating photorealistic red yarn textures as part of a multimodal creative workflow in Gemini Omni.
photorealistic red yarn
Author: Charly Wargnier | Source: Link | Published: May 28, 2026
A creative prompt used to generate a video of a whimsical hybrid creature merging a snail and a kitten.
Snail + Kitten = Snitten
Author: Build Fast with AI | Source: Link | Published: May 28, 2026
A technical prompt designed for Gemini Omni to generate a stable anime character walking animation with consistent hair, clothing, and colors.
Anime girl in a purple-haired dress walks steadily forward, with her hairstyle, hair color, dress style, and color remaining constant throughout, her figure and appearance unchanged, and her image across frames unified and stable.
Author: CloudwiseAI | Source: Link | Published: May 28, 2026
A prompt to generate high-speed FPV drone footage starting from a specific point on a map screenshot using Gemini Omni Flash.
High-speed FPV drone footage. Complete first-person view (pure FPV POV). The footage begins from the starting point on the screen.
Author: タナベ | 動画・音声生成AI解説 | Source: Link | Published: May 27, 2026
A prompt to generate a POV horse-riding virtual tour of Middle Earth locations based on an uploaded map image with direction markers.
Analyze the location and direction of the arrow on the attached map. remove the red arrow. Create a virtual tour guide video showing the 2 most important locations on that place.
POV: we were riding a brown horse.
Real gritty found footage.
[OPTIONAL] (race: elfs, orcs, hobbits...) doing everyday stuff
Author: TechHalla | Source: Link | Published: May 27, 2026
An assembly animation prompt that transforms a matte car frame into a finished glossy red sports car using isometric camera angles.
Create a fast-paced high-energy 10 sec video showing the rapid, piece-by-piece assembly of a sports car, using the provided 2 reference images as start to end points
- start with the four wheels rolling into position
- metallic rods and suspension quickly snaps into place b/w them
- engine block drops
- external body panels fly from all sides and snaps onto the frame
- transform matte primer into glossy red paint
keep the camera angle strictly as a 3/4 isometric shot
Author: Harshith | Source: Link | Published: May 27, 2026
A prompt capturing the bustling and nostalgic atmosphere of a Japanese shopping district from the Showa era with lively crowd sounds.
Showa era shopping street, bustling, Japanese language, lively
Author: Nobu_cyan | Source: Link | Published: May 27, 2026
A cinematic video generation featuring a city street at twilight, characterized by misty atmosphere and a movie-like aesthetic.
Lost in the misty twilight of the city where every street corner feels like a scene from a movie.
Author: Shah | Source: Link | Published: May 27, 2026
A poetic prompt for generating a nostalgic video of childhood memories illuminated by a soft glow during a rain shower.
Glowing childhood memories in rain
Author: Mr Das | Source: Link | Published: May 27, 2026
A video generation prompt that uses a storyboard image to create a 15-second anime-style cooking video of fluffy, jiggling souffle pancakes.
Based on the attached storyboard image, please create a 15-second horizontal 16:9 anime cooking video.
Theme: Extremely thick souffle pancakes that jiggle.
Follow the panel order of the storyboard and turn it into a video as a fast-paced cooking scene.
Author: ヴェルモット | Source: Link | Published: May 27, 2026
A highly detailed 10-second documentary prompt tracking human evolution from early primates to modern humans with seamless morphing and cinematic lighting.
Please create a 10-second vertical (9:16) ultra-realistic live-action documentary video using the attached storyboard image as an accurate visual blueprint. Strictly maintain the storyboard's composition, chronological order, human evolution, and walking direction. No extra shots or changes in sequence. Style: Ultra-realistic movie quality, high-budget documentary feel like BBC Earth or NHK Special, natural light, cinematic depth. Evolution: Transitions should feature seamless morphing of skeleton, face, and clothing while walking. Character: Walk from left to right at a constant speed, maintaining lineage continuity. Camera: Constant side-view tracking. Captions: Minimal documentary-style text in Japanese for each era. Negative: Slow motion, robotic movement, fantasy effects, low resolution.
Author: タナベ | 動画・音声生成AI解説 | Source: Link | Published: May 27, 2026
An image-to-video (i2v) prompt designed to simulate the physical action of a man throwing a bottle upward and having it land upright on a table.
He is a Japanese man in his 50s. Maintain consistency of the bottle shape until the end. Progress through the following sequence: 1. He throws the bottle held in his right hand upwards and completely lets go of it (after that, he absolutely does not touch the bottle). 2. The bottle rotates and falls, landing upright in its original position on the table. 3. He spreads his arms and rejoices.
Author: onoken1025 | Source: Link | Published: May 27, 2026
A professional video prompt for Gemini Omni focusing on precise typography placement, technical font styles, and synchronized narration effects.
White Helvetica lowercase text + thin line connecting to the main subject + inner monologue style narration
Author: AI Will | Source: Link | Published: May 27, 2026
A straightforward prompt for generating a game-style trailer featuring Indian cultural landmarks like the Taj Mahal.
generate a GTA India trailer
Author: Build Fast with AI | Source: Link | Published: May 27, 2026
A detailed cinematic prompt for generating a short travel film featuring an unfolding map and a destination ticket.
[CITY] = TOKYO
Create a 10-second cinematic travel memory film for [CITY].
A folded paper map and an old travel ticket lie on a clean ivory desk.
The ticket has the destination name “[CITY]” printed on it.
As soft light moves across the paper, the map slowly unfolds by itself.
Author: マグマグ | Source: Link | Published: May 27, 2026
A prompt to create a high-speed animation showcasing 20 different versions of a simple logo while keeping the design centered.
Rapid fire animation through 20 different versions of this logo. Keep it black on white and centered.
Author: Greenfield Team! | Source: Link | Published: May 27, 2026
A comprehensive prompt for generating a highly realistic documentary video showing the evolution of humanity from early primates to modern humans, maintaining consistent side-view walking motion and period-accurate backgrounds.
Please use the attached storyboard image as an exact visual blueprint to create a 10-second vertical (9:16) ultra-realistic live-action documentary video. Strictly maintain the storyboard composition, chronological order, human evolution, and walking direction. Adding extra shots or changing the sequence is prohibited.
--- [Visual Style] --- Ultra-realistic live-action film quality. High-budget documentary footage like BBC Earth, NHK Special, or National Geographic. Realistic texture like a historical recreation film. Natural light-based. Cinematic lighting. Photorealistic. Realistic skin, fur, clothing, stone tools, and natural environments. Cinematic depth. Intellectual and solemn atmosphere.
--- [Video Structure] --- Video length is 10 seconds. Switch through the following 6 scenes from the storyboard in order from 1 to 6 with equal length: 1. Ape-man (Early Primate), 2. Australopithecus, 3. Homo erectus, 4. Neanderthal, 5. Cro-Magnon, 6. Modern Human. Edit each scene to connect naturally. Emphasize the sense of gradual evolution over time.
--- [Character Direction] --- Characters walk from left to right in all scenes. Walking speed is always constant. Feel the continuity of the same lineage across all scenes. As evolution progresses, posture, facial features, physique, hair volume, clothing, tools, and walking style naturally evolve. Express evolution within the same lineage, not a switch to a different person.
--- [Camera] --- Maintain the same side-view composition in all scenes. Keep camera distance constant. Always show the character's full body. Smoothly track the walking motion. Violent camerawork is prohibited. Natural documentary-style filming.
--- [Background Direction] --- The background environment naturally evolves with each era: Forest, Grassland, Dry Zone, Glacial Zone, Mountainous Zone, Modern City. Ensure environmental changes connect naturally.
--- [Captions] --- Display short Japanese captions indicating the era in each scene. Content: Ape-man (Early Primate), Australopithecus, Homo erectus, Neanderthal, Cro-Magnon, Modern Human. Style: Small, elegant, non-intrusive, semi-transparent, white or light gray, centered at the bottom, modest fade-in/fade-out, BBC Earth/NHK style, Minimal elegant documentary-style typography. Important: Do not hide people with captions. No flashy animations. No YouTube-style UI. No variety show style.
--- [Music] --- Solemn and grand classical music evoking the beginning of the universe and human history. Intellectual and mysterious atmosphere. Emotional buildup like a movie trailer.
--- [Important] --- Strictly refer to the uploaded storyboard. Maintain frame layout, era changes, and walking direction. Do not show storyboard frames, annotations, UI, page info, or production notes in the final video.
--- [Negative Prompt] --- Slow motion, unnatural movement, dancing, exaggerated acting, robotic movement, unnatural muscles, deformed limbs, blurred face, low-quality texture, fantasy elements, sci-fi elements, mutation, violent camerawork, static background characters, low resolution, anime-style expression, comical expression, unnatural walking, camera angle change, person becomes a different person, excessive motion blur.
Author: タナベ | 動画・音声生成AI解説 | Source: Link | Published: May 27, 2026
Generates a video of a person reciting the months of the year with dynamic background changes corresponding to the seasons and holidays of each month.
A video of a man saying the months from Jan to Dec, one month every 1s. show the current month in the corner. No dialogue other than the month he says. The background location and scene changes every month to match it: Jan winter, feb valentine, march spring, April rain, May flowers, June beach, July fireworks, Aug tropical, Sept school/autumn, Oct Halloween, Nov cozy fall, Dec Christmas
Author: Harshith | Source: Link | Published: May 26, 2026
A detailed prompt for generating a video of a man counting to ten on his fingers, featuring specific instructions for timing, dialogue, and hand usage.
A video of a man counting to 10 on his fingers, show the number in the corner. A new number every 1s, no dialogue other than the numbers he says. He uses two hands for numbers bigger than 5.
Author: fofr | Source: Link | Published: May 26, 2026
This prompt transforms a photo of Tokyo Tower into a launching rocket, complete with a countdown and smoke effects at specific timestamps.
At second 0: A countdown voice in English is heard (from 3 to zero)
At second 4: When the countdown ends, smoke rises from under the central red tower (which is actually a rocket).
At second 5: The red tower slowly flies up into the sky.
Author: onoken1025 | Source: Link | Published: May 26, 2026
Contributions are welcome! Please feel free to submit a Pull Request.
- Fork the repository
- Add your prompt following the existing format
- Submit a PR
This work is licensed under CC BY 4.0. You are free to share and adapt, with appropriate credit.
- Google DeepMind for developing Gemini Omni
- YouMind for the prompt gallery and community
- All prompt contributors
📝 Submit a Prompt • ⭐ Star this repo
🤖 This README is automatically generated. Last updated: 2026-06-01T07:31:25.855Z