The Gemini Omni 1.2 Flash prompt generator gives you 20 free, copy-ready video prompts for Google's new #1 text-to-video AI. Conversational multi-turn editing, 15-second clips, and native synchronized audio — all in one model.
The Gemini Omni 1.2 Flash prompt generator on this page provides 20 free, professionally crafted prompts for Gemini Omni 1.2 Flash — Google's AI video generation and editing model that reached general availability on November 5, 2026. Omni 1.2 Flash currently ranks #1 on the Arena text-to-video leaderboard with an Elo score of 1540, making it the strongest publicly available AI video model by human preference voting.
Gemini Omni 1.2 Flash generates video clips of up to 15 seconds from a single text prompt, with native synchronized audio generated in the same pass. Its defining capability is conversational multi-turn editing: you can generate a clip, then follow up with instructions like "now remove the background figures" or "make the lighting warmer" and the model applies your edit to the same scene. No other public model handles this as reliably.
Every prompt below is structured for Omni 1.2 Flash's strengths — narrative scene description, explicit audio direction, camera movement, and clear duration. Paste directly into Google AI Studio, the Gemini API, or the Gemini app.
Omni 1.2 Flash parses full narrative scene descriptions. Use this structure:
Click any prompt to copy — paste into Google AI Studio, the Gemini API, or the Gemini app
A cinematic arrival sequence: a traveller steps off a train onto a sunlit platform in Kyoto, Japan. The camera follows at mid-distance — platform receding behind, the traveller slowing to take in the scene ahead. Cherry blossom petals drift across the frame. 12 seconds, golden-hour light, 24fps. Audio: the train departing behind them, birdsong ahead, a distant temple bell. Quality: theatrical feature film.
A luxury commercial shot: a glass perfume bottle rises from absolute black, illuminated by a single amber strip light from below. It rotates one full revolution over 10 seconds, the liquid inside shifting and catching the light. At the 10-second mark, the light expands to reveal the full bottle surrounded by floating dried petals. 12 seconds total. Audio: a single sustained cello note building through the reveal. Studio quality, 24fps.
A fashion editorial shot in a minimal white studio: a model in an oversized ivory trench coat walks directly toward the camera from 10 metres away, pace deliberate, expression composed. The camera stays fixed. At 3 metres out, she stops and holds. Light source: single softbox camera-left creating a clean shadow across the floor. 12 seconds. Audio: the soft percussion of footsteps on polished concrete, then silence. Vogue editorial quality.
A wildlife telephoto shot: a polar bear wading through shallow Arctic water at low tide, turning its head toward the camera at the 7-second mark. The bear shakes water from its fur, then continues walking out of frame. Foreground: shallow water ripples. Background: pack ice and low grey sky. 13 seconds. Audio: the crunch and splash of footsteps in water, wind over ice, distant seabirds. BBC Planet Earth quality, natural light.
Start with a wide shot of a modern café interior — round tables, pendant lights, a few customers. Over 8 seconds, progressively clear the frame: tables and customers fade from the edges inward until the final 4 seconds show the empty room with morning light pouring through the windows. The transition should feel like a time-lapse emptying, not a hard cut. 12 seconds. Audio: café ambient sound fading to quiet room tone and distant traffic.
An architectural reveal at dawn: the camera begins at street level looking up at a contemporary glass-and-steel library facade, then slowly rises alongside it — each floor's grid of windows lighting up as dawn breaks — until at the top the camera pulls back to show the full building against a rose-gold sky. 14 seconds. Audio: the ambient sound of a city waking — distant traffic, a bus, birdsong at the top of the ascent. Architectural photography quality.
A street documentary sequence at a night market in Taipei: Shot 1 (4s) — wide establishing shot, vendor stalls lit in orange and neon, crowds moving through; Shot 2 (4s) — close on a vendor's hands lifting skewers from a grill, steam rising; Shot 3 (5s) — a couple sharing food at a wooden stall, laughing. Continuous ambient audio: the sizzle of the grill, overlapping conversations, music from a nearby speaker. Documentary quality, natural light + neon.
A surf documentary sequence: a surfer drops into a head-high wave at golden hour, bottom-turns, executes a single clean top-turn, then pulls off the back as the wave closes out. The camera is positioned at water level, slightly behind the surfer. Spray catches the low sun. 12 seconds. Audio: the rush of water, the crack of the board on the top-turn, receding wave sound. Surf film quality, warm colour grade.
A music video sequence: a dancer in a black outfit moves through a neon-lit corridor, the camera tracking laterally to keep pace. The choreography is sharp, isolating — arms, chest, shoulders — with one full-body spin at the 8-second mark. Neon tubes line the ceiling: pink, cyan, amber. Floor: polished black. 12 seconds. Audio: a minimal synth beat with a sharp high-hat on every step, building to the spin. Visual quality: polished R&B music video.
A sci-fi interior sequence: a technician in a worn grey jumpsuit moves through a narrow space station corridor, checking wall panels. Through a circular porthole at the corridor's end, a gas giant fills the view. The technician pauses at the porthole for 5 seconds, then pushes off and continues out of frame. Camera: handheld-style tracking shot transitioning to static at the porthole. 14 seconds. Audio: the hum of life support, a soft metallic tap on the porthole frame. Arrival-level production design.
A timelapse sequence of a dramatic weather change over open farmland: clear blue sky compresses to heavy grey cloud cover in 10 seconds, lightning visible at the horizon in the final 3 seconds. Camera: static wide angle, slightly elevated, farmland filling the lower third. The light shifts from warm afternoon sun to flat grey as the storm approaches. Audio: birdsong fading, wind rising, the first low thunder at 11 seconds. 13 seconds total. National Geographic quality.
A three-shot cooking sequence: Shot 1 (4s) — fresh pasta sheets being cut on a floured wooden board, close-up on the knife and the folded dough; Shot 2 (4s) — pasta lifted from boiling water and tossed in a pan with sauce, steam rising dramatically; Shot 3 (5s) — the finished dish plated in a white bowl, fork twirling. Warm under-counter LEDs throughout. 13 seconds total. Audio: the slice of the knife, the bubble of the water, the sizzle of the pan, the clink of the fork. Restaurant documentary quality.
A studio portrait sequence demonstrating light transitions: a subject stands before a seamless grey background. The key light shifts position over 10 seconds — from frontal (flat, approachable) to 45° left (Rembrandt lighting, authoritative) to pure side-lighting (dramatic). The subject remains still throughout, expression composed. Final 2 seconds: hold on the side-lit frame. Audio: studio silence. 12 seconds. Peter Lindbergh-quality studio photography.
A travel documentary shot through a narrow street in a Moroccan medina: the camera walks at shoulder height through a tunnel of shade — colourful ceramics and hanging lanterns on either side — emerging into a bright courtyard with a central fountain at the 10-second mark. The transition from shade to light is deliberate. 14 seconds. Audio: distant call to prayer, the echo of footsteps on stone, the sound of the fountain arriving as we exit the alley.
An ink animation sequence in classical East Asian brush style: a single Chinese character for 'mountain' (山) is painted stroke by stroke in real-time — the brush enters the frame from the top, the first stroke descends firmly, then the two flanking strokes in sequence. The animation captures the pressure and release of each stroke. Black ink on warm rice paper. 10 seconds, slow and deliberate. Audio: the whisper of the brush on paper, complete silence otherwise.
A marathon finish-line documentary sequence: Shot 1 (5s) — a runner approaching the finish line in a crowd of spectators, expression showing the final push; Shot 2 (4s) — the runner crossing, arms raised; Shot 3 (4s) — the runner stopping beyond the line, hands on knees, exhaling deeply. Camera static for Shot 1, cuts to tracking at Shot 2, holds on Shot 3. Audio: crowd cheering building to peak at the cross, then dispersing to individual sounds of exertion and relief. 13 seconds.
An automotive commercial sequence: a sleek sedan drives along a wet urban road at night — city lights reflecting in long streaks on the tarmac. The car passes beneath a series of overhead lights in slow motion, water spraying in controlled arcs from the rear wheels. Camera: low side angle, tracking at car speed. 12 seconds. Audio: the controlled hiss of wet tyres, rain drops on the roof, ambient city sound underneath. Premium automotive quality.
An interior design reveal sequence: a plain white room with a single chair transforms over 10 seconds as furnishings, light, and colour appear in sequence — a rug materialises, then a sofa, then warm pendant lighting activates, then bookshelves fill. Each element appears with a soft fade-in. The final 3 seconds hold on the completed room, warm evening light from an unseen window. Audio: each addition accompanied by a subtle layering of ambient home sounds — fabric settling, light hum of fixtures. 13 seconds.
A drone reveal from cliff base to sea horizon: the camera begins at the base of dramatic chalk cliffs, slowly rising alongside the face — seabirds visible on ledges — until it clears the cliff top and the full expanse of open sea fills the frame, meeting the horizon. 14 seconds. Audio: waves crashing at the base fading to open-sea wind as we rise. The transition from enclosed rock face to open sky should feel like emergence. Cinematic aerial photography quality.
A forest at dusk: tall pine trees, shafts of late light filtering through the canopy, a deer visible at the treeline 30 metres in. The deer holds still for 5 seconds, then turns and walks deeper into the forest. The camera stays static on the treeline as the deer disappears. 13 seconds. Audio: forest ambient — distant birdsong, a light breeze in the upper canopy, the soft crack of a twig as the deer moves. Golden-hour light. BBC natural history quality.
| Model | T2V Rank | Max Length | Audio | Best For |
|---|---|---|---|---|
| Gemini Omni 1.2 Flash Google | #1 T2V (Arena 1540) | Up to 15 seconds | Native synchronized | T2V quality, conversational editing |
| Kling 4.0 Kuaishou | Top 5 T2V | Up to 30 seconds | Native audio | Cinematic motion, long clips |
| MiniMax H3 MiniMax | #2 I2V (Elo 1351) | Up to 15 seconds | Native stereo | Multi-reference, 2K resolution |
| Dreamina Seedance 2.5 ByteDance | #2 T2V (Arena 1493) | Up to 30 seconds | Native audio | Lip-sync, dialogue, long clips |
| Wan 3.0 Alibaba | Top open-source | Up to 30 seconds | Audio support | Open-source, local deployment |
Arena leaderboard rankings as of November 2026. Rankings update continuously — check arena.ai for the latest.
The Gemini Omni 1.2 Flash prompt generator on this page provides 20 free, copy-ready video prompts for Google's Gemini Omni 1.2 Flash — the AI video and editing model that reached general availability on November 5, 2026. Gemini Omni 1.2 Flash currently ranks #1 on the Arena text-to-video leaderboard (Elo 1540), making it the strongest publicly available AI video model. Each prompt is structured for Omni 1.2 Flash's strengths: conversational multi-turn editing, native synchronized audio, and 15-second clip generation.
Gemini Omni 1.2 Flash is Google's latest AI video and multimodal generation model, released November 5, 2026. Key upgrades from 1.1 Flash: (1) Maximum clip length extended from 10 seconds to 15 seconds per generation; (2) Pricing reduced from approximately $0.10/s to approximately $0.07/s at 720p — a 30% cost reduction; (3) Improved motion consistency and temporal coherence across longer clips; (4) Better conversational editing — following up a generation with specific edit instructions works more reliably in 1.2 than 1.1. From December 3, 2026, requests to the Gemini Omni 1.1 Flash API will route automatically to 1.2 at 1.2 pricing.
Gemini Omni 1.2 Flash responds to narrative scene descriptions. Use this structure: (1) Camera angle and movement — 'static wide angle', 'low side tracking shot', 'drone ascending'; (2) Subject and action with timing — describe what happens and in what sequence; (3) Light source and quality — 'golden-hour backlight', 'single softbox camera-left', 'neon corridor'; (4) Audio intent — Omni 1.2 Flash generates synchronized audio natively, so describe the sounds you want; (5) Duration — keep it within 15 seconds; (6) Quality benchmark — 'BBC documentary quality', 'Vogue editorial', 'theatrical feature film'. Write in complete descriptive sentences, not keyword tags.
Gemini Omni 1.2 Flash is accessible via Google AI Studio (aistudio.google.com) with a free tier, the Gemini API for developers, and the Gemini app for consumers. Third-party platforms including Runway and Artlist have also integrated Gemini Omni Flash. The model ID in the API is listed on the Google AI developer docs. For multi-turn conversational editing — generating a clip then following up with 'now remove the background figures' — the API and AI Studio both support this natively in Omni 1.2 Flash.
Gemini Omni 1.2 Flash, Kling 4.0, and MiniMax H3 are the three strongest AI video models as of November 2026. Gemini Omni 1.2 Flash leads on the Arena text-to-video leaderboard (Elo 1540) and is the best choice for conversational multi-turn video editing. Kling 4.0 excels at cinematic camera motion, long-subject consistency, and 30-second clips with 10 keyframes. MiniMax H3 leads on image-to-video (I2V) and has the strongest multi-reference input system (9 images + 3 videos + 3 audio clips). For quick text-to-video generation with the strongest automated results, Gemini Omni 1.2 Flash is the current leader.
Gemini Omni 1.2 Flash has a free tier through Google AI Studio with limited daily generations. Beyond the free quota, it costs approximately $0.07 per second of 720p video generated — a 15-second clip at standard quality costs approximately $1.05. API pricing is listed on the Google AI developer pricing page. The Gemini app (consumer) includes Omni Flash video generation with a Gemini Advanced subscription (Google One AI Premium). No signup is required to view and copy the prompts on this page.
More free generators in this collection — no signup, unlimited use.
20 free prompts for Kling 2.5 Turbo — #1 on Artificial Analysis video leaderboard, fastest generation.
20 copy-paste prompts for Kling 3 — the creator's workhorse for AI video in 2026.
20 free prompts for Google Veo 4 — cinematic 4K, native audio, up to 30 seconds.
20 copy-paste prompts for HappyHorse — Alibaba's #1-ranked AI video model globally.
20 free prompts for SkyReels V4 — first open-source video model with synchronized audio.
20 copy-paste prompts for ByteDance Seedance 2.5 — dialogue lip-sync, native audio.