Gemini Video Prompts You Can Copy-Paste for Reels (India, 2026)

Rahul
1 November 2026LinkedIn
Gemini Video Prompts You Can Copy-Paste for Reels (India, 2026)

To make a Reel with Gemini, open Create video in the Gemini app, paste a prompt that names the shot, subject, action, lighting and any on-screen text, and upload a vertical image if you want a portrait video. Google's own page says Gemini now uses the Omni model for 10-second clips with native audio.

If you have been searching for a Gemini video prompt to copy and paste, check the model name first. Many guides still describe Veo and the limits from 2025. Google's Gemini video page now says Gemini Omni 1.1 Flash replaces the earlier Veo 3.1 model in the app, and the prompting advice is different. This post gives you ten templates for Indian creators and small businesses. For prompts that work in Google Flow, Runway, Kling and Luma, see our AI video prompts copy-paste guide for India, and for Flow's free options read our Google Flow AI video generator tutorial.

One honest note before we start: the templates below are our own, built from Google's published guidance. We have not tested them in your account, and video models give different results each time. Treat them as starting points.

What changed: Gemini now uses Omni, not Veo

Google announced Gemini Omni at I/O on 19 May 2026, starting with video. Engadget and Decrypt both reported the launch that day. The first model, Omni Flash, works in the Gemini app, Google Flow and YouTube Shorts for Google AI Plus, Pro and Ultra subscribers.

What Google's pages say today (read on 1 November 2026):

Feature

What Google says

Clip length

"Create 10 second videos" (Gemini video page)

Sound

Native audio generation

Inputs

Text, plus up to 5 images and one video (Gemini Help)

Editing

Edit by conversation, e.g. change a background or add a caption

Longer videos

Scene extension adds footage in 10-second steps, up to 40 seconds total (Google blog, 27 August 2026)

Plan needed

A Google AI plan for personal accounts; features "vary by tier and geography"

Age

Not available to users under 18

Watermark

Videos carry Google's SynthID watermark (Engadget)

The last dated source I could find is from 27 August 2026, so nothing in the last 30 days. The live Gemini page is undated. Check it again before you rely on any detail.

Plans and price

Google's subscription page opened for the United States on 1 November 2026, so those figures are not India prices. It showed Google AI Plus at $4.99 a month and AI Pro at $19.99 a month, with higher usage limits than the free plan, and Ultra starting at $99.99. The Indian rupee price is not shown there, so open the Gemini app or the Google AI plans page in India to see your own price. I am not quoting any daily video limit, because the current official pages I read do not state one. Older articles that say "3 videos per day" describe the 2025 setup.

The Gemini video prompt formula from Google's own guide

Google DeepMind published a prompt guide for Omni. Its main advice is simple: "The more detail you add, the more control you'll have over the final output." It breaks a prompt into five parts:

  1. Shot and camera: framing and movement, such as static, push in, or a natural smartphone zoom.
  2. Style: realistic, cinematic, anime, claymation, watercolour.
  3. Lighting: the light source and quality, such as crisp, warm or soft.
  4. Location: the setting in broad terms.
  5. Action: who or what is in the scene and how they move.

Two more tips from the guide are useful for Reels. For an unbroken take, ask for "one continuous shot". For on-screen text, you can specify how it appears, including word by word. Google also says Omni, unlike Veo, does not need highly prescriptive prompts, so you can state your goal and let the model fill in details.

How to get a vertical (9:16) video for Reels

Gemini's help page says that for text prompts the default is landscape, and if you upload a reference image or video, the result matches the uploaded file's aspect ratio. I did not find a documented way to request portrait in the text itself, so:

  1. Upload a vertical photo or screenshot as your reference (a portrait phone photo works).
  2. Write your prompt describing what should happen.
  3. Check the preview is portrait before you generate more.

If portrait does not come out the way you want, crop in your editor. Test one clip first.

10 full copy-paste Gemini video prompts for Reels

These are complete, long prompts, not one-liners. Google's prompt guide says "The more detail you add, the more control you'll have over the final output", so each prompt below spells out the shot, subject, action, location, lighting, style, sound, on-screen text and a 10-second timeline.

How to use them:

  1. Upload a vertical (portrait) reference image first if you want a 9:16 Reel (see the section above).
  2. Copy the whole block, including every line, into Gemini's Create video box.
  3. Replace everything in [square brackets] with your own details. Delete any line you do not need.
  4. Generate once, watch it fully, then edit by conversation ("keep everything the same, but...") instead of rewriting the whole prompt.

These are our templates, built from Google's published guidance. We have not tested them in your account, and results vary between runs.

1. Gemini video prompt for a product ad (chai or snack brand)

Create a 10-second vertical product ad for an Indian snack brand.

SHOT: One continuous shot, no jump cuts. Start as a close-up and slowly push in, ending on the product packet.
SUBJECT: A steel glass of hot masala chai and a packet of [brand name] [product, e.g. jeera biscuits] on a worn wooden table.
ACTION: 0-3 seconds: steam curls up from the chai. 3-6 seconds: a hand with a silver ring dips a biscuit into the chai and lifts it out. 6-10 seconds: the hand places the packet upright beside the glass, logo facing camera.
LOCATION: A small sunlit kitchen in an Indian home, steel utensils softly blurred in the background.
LIGHTING: Warm early-morning sunlight from a side window, soft shadows, golden highlights on the steam.
STYLE: Realistic, cinematic food commercial, shallow depth of field, rich warm colours.
SOUND: Gentle clink of a steel spoon, soft morning birds outside, light acoustic guitar music.
ON-SCREEN TEXT: At 7 seconds, fade in at the bottom centre in bold white letters: "[tagline, e.g. Har chai ka saathi]". Keep the text on screen until the end.
KEEP IT CLEAN: No other brand logos, no people's faces, no extra text.

2. Gemini video prompt for festive sweets (Diwali or wedding season)

Create a 10-second vertical festive Reel for a sweet shop.

SHOT: Static shot, locked off, slightly above table height.
SUBJECT: A brass thali filled with [mithai names, e.g. kaju katli, motichoor laddoo, gulab jamun], marigold flowers and lit clay diyas around it.
ACTION: 0-3 seconds: diya flames flicker gently. 3-7 seconds: several hands of different ages reach in from both sides and each picks up a sweet. 7-10 seconds: the thali is shown again with a few sweets gone, sparklers glowing softly in the blurred background.
LOCATION: A decorated Indian living room with rangoli on the floor and string lights on the wall.
LIGHTING: Warm diya light and fairy lights, deep golden tones, soft glow on the sweets.
STYLE: Realistic, warm, festive, cinematic.
SOUND: Laughter and happy family chatter, soft shehnai-style festive music.
ON-SCREEN TEXT: At 6 seconds, top centre in elegant gold font: "[Shop name]". At 8 seconds, bottom centre: "Order now: [phone or area]".
KEEP IT CLEAN: No faces in close-up, no fireworks in frame.

3. Street food close-up

Create a 10-second vertical street food Reel.

SHOT: One continuous shot. Start in an extreme close-up of a hot tawa, then a natural smartphone zoom out to show the stall.
SUBJECT: A street cook making [street food item, e.g. pav bhaji] on a large iron tawa.
ACTION: 0-3 seconds: butter melts and sizzles on the tawa. 3-6 seconds: the cook mashes and flips the food quickly with a flat spatula. 6-10 seconds: zoom out to reveal the stall with its hand-painted name board and a small queue of customers.
LOCATION: A busy evening street in [city, e.g. Mumbai], traffic lights and shop signs blurred behind the stall.
LIGHTING: Warm bulb light at the stall mixed with cool blue evening light from the street.
STYLE: Handheld, energetic, realistic phone-video look.
SOUND: Loud sizzling, the clang of the spatula, street traffic and the vendor calling out orders.
ON-SCREEN TEXT: Word by word, one word on the screen at a time, starting at 2 seconds: "Best [food item] in [area]?"
KEEP IT CLEAN: No other brand logos, no text other than the line above.

4. Real estate walkthrough

Create a 10-second vertical property walkthrough Reel.

SHOT: One continuous shot. A smooth, gliding camera moves forward at walking pace through the home.
SUBJECT: A bright, furnished [2BHK or 3BHK] apartment in [city].
ACTION: 0-3 seconds: the camera enters through the open main door into the living room. 3-7 seconds: it glides past a sofa and dining table towards the balcony. 7-10 seconds: it stops at the balcony railing, showing an open city or green view.
LOCATION: Modern Indian apartment with light wooden floors, white walls, indoor plants and large windows.
LIGHTING: Soft natural daylight filling the room, clean and airy.
STYLE: Realistic, premium real-estate video, wide lens, steady.
SOUND: Calm piano music, faint birds from the balcony.
ON-SCREEN TEXT: At 1 second, top left: "[Project name], [locality]". At 8 seconds, bottom centre in bold: "[Configuration] from [price, if you are allowed to show it]".
KEEP IT CLEAN: No people, no clutter, no visible brand names on furniture.

5. Fashion or boutique Reel

Create a 10-second vertical fashion Reel for a boutique.

SHOT: Film camera look. Start as a medium shot, then a slow push in to a close-up of the fabric.
SUBJECT: A woman in her late 20s wearing a [colour] [saree, lehenga or kurta set] with [embroidery type, e.g. zari border].
ACTION: 0-4 seconds: she turns slowly once in place, fabric flowing. 4-7 seconds: she lifts the pallu or dupatta gently towards camera. 7-10 seconds: close-up of the embroidery catching the light.
LOCATION: A boutique with warm wooden shelves, folded fabrics and a full-length mirror.
LIGHTING: Warm tungsten light with soft highlights on the fabric.
STYLE: Elegant, cinematic, soft slow motion, rich colours.
SOUND: Soft sitar and tabla music.
ON-SCREEN TEXT: At 7 seconds, bottom centre in a thin serif font: "[Boutique name] | [collection name]".
KEEP IT CLEAN: Modest, tasteful styling. No other brand labels visible.

6. Gemini video prompt for an explainer hook (educator or coach)

Create a 10-second vertical talking-head hook Reel.

SHOT: Webcam style, medium close-up, the person centred and looking straight into the camera.
SUBJECT: A friendly Indian [teacher or coach] in their 30s, wearing a plain [colour] shirt, sitting at a tidy desk.
ACTION: 0-4 seconds: they lean slightly forward and say: "[your hook line, e.g. 90% students make this one mistake in exams]". 4-8 seconds: simple animated icons of [topic, e.g. a clock, a notebook, a red cross] pop up one by one beside them. 8-10 seconds: they smile and point down towards the caption.
LOCATION: A bright home study with a bookshelf softly blurred behind.
LIGHTING: Clean, even, bright daylight from the front.
STYLE: Simple, modern, clean social-media look.
SOUND: The speaker's clear voice in [Hindi or English], light background beat.
ON-SCREEN TEXT: Word by word, synced to the speech, in bold yellow captions at the bottom.
KEEP IT CLEAN: No logos, no extra text beyond the captions.

7. Travel or heritage Reel

Create a 10-second vertical travel Reel.

SHOT: One continuous shot. A slow push in from a wide view towards [monument, fort or beach].
SUBJECT: [Place name] at golden hour.
ACTION: 0-4 seconds: a wide view with a few visitors walking in the foreground. 4-7 seconds: a flock of birds crosses the sky above the monument. 7-10 seconds: the camera settles on the main structure as the sun dips behind it.
LOCATION: [Place name, city, state], India.
LIGHTING: Warm golden-hour sunlight, long soft shadows, glowing sky.
STYLE: Realistic, cinematic travel film, calm pace.
SOUND: Light breeze, distant temple bells or waves (choose one), soft instrumental music.
ON-SCREEN TEXT: At 6 seconds, centre in clean white letters: "[Place name] | Best time: [season]".
KEEP IT CLEAN: No close-up faces of visitors, no added logos.

8. Gemini video prompt for photo-to-video (upload 1 to 5 product photos)

Use my uploaded photos of [product]. Create a 10-second vertical product Reel from them.

SHOT: Keep the framing of each photo. Move between photos with smooth camera motion, not hard cuts.
ACTION: 0-3 seconds: bring the first photo to life with a slow push in and soft movement in the background. 3-6 seconds: glide to the second photo with a gentle sideways sweep. 6-9 seconds: show the third photo with a slow zoom out. 9-10 seconds: end on the clearest product shot.
STYLE: Keep the colours, shapes and labels exactly as they appear in my photos. Clean, modern, realistic.
LIGHTING: Match the lighting already in the photos.
SOUND: Upbeat, light music that suits a [category, e.g. skincare] brand.
ON-SCREEN TEXT: At 8 seconds, bottom centre: "[Product name] | [price or offer]".
KEEP IT CLEAN: Do not change the product design, label text or colours. Do not add other products.

9. Gemini video prompt to edit an existing clip

Google's guide says short, specific edit requests work best. Upload your clip, then use one of these, one at a time:

Edit this keeping everything the same. Change only the background to [new background, e.g. a bright modern cafe]. Keep the person, their clothes, their movement and the audio exactly as they are.
Edit this keeping everything the same. Add the caption "[caption text]" at the bottom centre in bold white letters from the start to the end of the video.
Edit this keeping everything the same. Change the camera angle to a close-up that slowly widens to a medium shot over the full 10 seconds.

10. Story in 10 seconds

Entire story in 10 seconds. Create a vertical mini-story Reel for a small business.

SHOT: Handheld feel, natural movement, three clear beats in one continuous shot.
SUBJECT: A young woman who runs a small [shop type, e.g. bakery] in [town].
ACTION: 0-3 seconds: at dawn she lifts the shutter of her shop, yawning, keys in hand. 3-7 seconds: the shop fills up, customers smile as she hands over [product]. 7-10 seconds: evening, she sits on the shop step counting the day's orders on her phone and smiles.
LOCATION: A narrow market street in a small Indian town.
LIGHTING: Cool blue dawn light at the start, bright daylight in the middle, warm orange evening light at the end.
STYLE: Realistic, warm, hopeful, documentary feel.
SOUND: Shutter rolling up, market chatter, soft hopeful music building towards the end.
ON-SCREEN TEXT: At 8 seconds, centre in handwritten-style font: "[Your brand message, e.g. Har din ek nayi shuruaat]".
KEEP IT CLEAN: No other brand names on signboards.

Tip: if a long prompt gives you something busy or confused, cut it back. Keep SHOT, SUBJECT, ACTION and ON-SCREEN TEXT, and remove the rest. Then add details back one line at a time.

Mistakes that waste your credits and time

  • Writing a paragraph of unrelated ideas. One clip, one idea.
  • Skipping the camera. Say static, push in or smartphone zoom, or you leave the framing to chance.
  • Forgetting sound. Native audio is on, so say what you want to hear.
  • Asking for real people. Use only images you have the rights to, and never make videos meant to deceive. Google's help page says not to "generate or share videos to deceive, harass, or harm others".
  • Posting without a check. Watch the clip for odd hands, wrong text or brand mistakes before it goes live.
  • Skipping a plan for the whole Reel. A 10-second clip is one beat. For a 30-second Reel, write three prompts that tell one story, or use scene extension where your plan offers it, and edit them together.
  • Ignoring your audience. Decide the one thing a viewer should do after the clip, such as visit, call or save, and put that in the on-screen text of the last second.

Turn clips into money

A good prompt is one step. The harder part is turning clips into an offer: a short ad for a local shop, a Reel package for a client, or a content calendar. Zero To AI's AI Revenue System covers AI video, ads and earning. On 1 November 2026 its page lists it at ₹1,999, with 12 hours across 36 lessons, self-paced, with downloadable prompt templates and lifetime access. The page does not mention a certificate.

Sources

Last verified: 1 November 2026. Models, limits and prices change fast; confirm in the Gemini app.

Hands-on course
Build the automation, don't just read about it.

Learn to build AI workflows that handle your busywork — live sessions, real projects, zero code.

See the course

Beginner-friendly

Frequently asked questions

Which model does the Gemini app use for video now?

Google's Gemini video page says Gemini Omni 1.1 Flash replaces the previous Veo 3.1 model. Omni was announced on 19 May 2026. Because plans and models change, check the Gemini app for the exact model your account uses.

How long can a Gemini video be?

Google's page says you can create 10-second videos. The 27 August 2026 Google blog says scene extension adds footage in 10-second steps up to 40 seconds in total, for Plus, Pro and Ultra subscribers. Availability can vary, so test it in your account.

Is Gemini video generation free in India?

A Google AI plan is required for personal accounts, and features vary by tier and geography. I could not find an official India price or a free tier on the pages I read. College students can get a free Pro plan for one year, per the Gemini video page. Check the app for your options.

How do I make a vertical video for Instagram Reels?

Text prompts default to landscape, but uploads match the aspect ratio of your reference file. Start from a vertical image or clip. I did not find a documented text setting for portrait, so test a short clip first.

Can I use my own photos?

Yes. You can upload one video and up to five images, but only if you have the rights to them. The feature is not available to users under 18. Uploading videos for edits is not available in the EEA, Switzerland, the United Kingdom and some US states.

Comments

Loading comments…

Leave a comment

Related articles

You may also like these

4,000+ students enrolled

Reading about automation
won’t automate anything.

Build your first working AI agent this week — no code, no developer.

₹1,499₹4,999one-time
Start for ₹1,499Start for ₹1,499

Talk to a mentor
before you start

Not sure which course fits your goals? Our team will review where you are, recommend the right path, and answer every question, so you start with total confidence.

ZERO TO AI
© 2026 Zero to AI — All rights reserved.