Prompts
Instagram Reels AI Video Prompts: 10 Copy-Ready Prompts (2026)
Ten copy-ready Instagram Reels AI video prompts for 9:16 vertical on muvi.video, built for Veo 3.1, Seedance 2.0, and Kling 3.0 with aesthetic, trend-aware framing.
What Makes a Reel-Ready Prompt and Which Model Delivers It on muvi.video
A Reel that performs is not a horizontal clip cropped to fit. The vertical 9:16 frame, the aesthetic-first composition, and the music-friendly pacing are decisions you make before the model renders a frame. On muvi.video, Veo 3.1 is the default for Reels-format video, it produces clips of 4, 6, or 8 seconds at 1080p in native 9:16, with ambient and dialogue audio generated alongside the video, and it returns 4 variants per run so you can pick the strongest take. When a Reel needs a continuous beat longer than 8 seconds, Seedance 2.0 covers any integer duration from 4 to 15 seconds in native 9:16. Kling 3.0 handles stylized motion when your reference look is a specific texture rather than a technical spec. Pricing and coin costs vary by plan, see the pricing page for current details on what each tier includes. This page explains the anatomy of a Reels prompt and gives you ten copy-ready prompts written for vertical video that you can run today.
The 7 Components Every Reels Prompt Needs
A Reels AI video prompt is a layered specification, not a caption. Veo 3.1, and modern video models in general, respond to explicit framing, lighting, and motion language. Generic words like "aesthetic" or "viral" add noise. The seven components below map the decisions that separate a scroll-stopping vertical clip from a generic one.
1. Vertical Framing (9:16) Aspect ratio is a first-sentence decision. State "9:16 vertical, full-bleed subject, safe margins for caption and UI" before any other instruction so the model composes for the phone screen rather than cropping a wide frame. Keep the focal subject in the upper two thirds, the bottom strip is covered by the caption and audio controls in the Reels player.
2. Aesthetic and Mood Name the visual register precisely. "Warm film-grain aesthetic, soft pastel palette, gentle bloom on highlights" is concrete. "Cool desaturated street look, high contrast, crisp shadows" sets a different register. The aspirational, polished feel that performs on Reels comes from a named palette plus consistent light, not from the word "aesthetic" alone.
3. Subject and Action State who or what is on screen and the single action they perform in the clip. "A barista pours latte art into a ceramic cup, slow and deliberate" gives the model one clear beat. One clip should carry one action, multi-action prompts in a short vertical clip read as rushed.
4. Camera Movement Specify a single movement with start and end. "Slow push in from medium to close-up over the clip" is actionable. "Dynamic camera" is not. Reels-friendly moves: slow push in, gentle handheld drift, locked-off with subject motion, and a top-down reveal for product or food.
5. Lighting Name the source and quality. "Soft window light from camera left, warm fill, no harsh shadow" reads polished and aspirational. "Hard practical neon from behind, cool rim light" reads editorial. Lighting is the highest-leverage cue for the premium look that lifts a Reel above generic stock.
6. Pacing and Music-Friendliness Reels are watched with audio and trend tracks. Build prompts that hit a clear visual beat (a reveal, a turn, a pour) the viewer can sync to a music drop. State "single clean beat, motion lands in the final second" so the clip has an editable sync point, even though you add the trend track in post.
7. Audio Intent Veo 3.1 generates native ambient and dialogue audio alongside the video. Decide upfront whether you want ambient sound (cafe tone, city air, fabric rustle) captured natively, or a silent clip you score with a trend track in the Reels editor. State "ambient cafe audio, no dialogue" or "no audio, clip scored in post" so the model does not add unwanted sound.
Which muvi.video Model to Use for Reels Video
Not every Reel needs the same model. The tree below maps the most common Reels use cases to the model that handles them best. Use it before writing your prompt, the wrong model costs coins and produces output a prompt rewrite cannot fix.
Veo 3.1, Best for: polished 9:16 clips, native audio, quick variant selection
Veo 3.1 generates clips of 4, 6, or 8 seconds at 1080p, and supports both 16:9 and 9:16, so it composes natively for the vertical Reels frame rather than cropping. It returns 4 variants per generation, useful for picking the strongest take of a hook or reveal. Native ambient and dialogue audio generates alongside the video (product claim per Google, verify current behavior on the platform). Veo 3.1 does not support deterministic seed control, each run with the same prompt may vary, so for a multi-clip Reel series rely on consistent subject, wardrobe, and setting descriptions, or anchor frames with the Veo 3.1 Image-to-Video variant. Standard and Fast variants are available, Standard for final delivery, Fast for cheap iteration. For current pricing and coin costs, see the pricing page. For detailed capabilities, see the Veo 3.1 model page.
Use Veo 3.1 when: you need a polished 9:16 hook or beat at 1080p, your clip is 8 seconds or shorter, you want native audio, or you want 4 variants to choose from.
Seedance 2.0, Best for: longer continuous Reels beats (up to 15 seconds)
Seedance 2.0 generates clips at any integer duration from 4 to 15 seconds, the widest range in the catalog. It supports six aspect ratios including 9:16, and outputs at 480p or 720p (the Start/End Frame variant outputs up to 1440p). It also offers omni-reference for carrying a look or subject across generations. When a Reel needs one continuous beat longer than Veo 3.1's 8-second ceiling, a walk-through, a long pour, a slow product turn, Seedance is the route. Resolution is lower than Veo 3.1, so reserve it for clips where duration matters more than pixel-level fidelity.
Use Seedance 2.0 when: your Reel needs a single continuous clip longer than 8 seconds, you want any specific integer duration from 4 to 15 seconds, or you want omni-reference to keep a look consistent across takes.
Kling 3.0, Best for: stylized motion aesthetics
Kling 3.0 has three variants in the catalog (Text-to-Video, Image-to-Video, Kling O3) and returns 3 variants per run. It handles stylized motion textures that other models render differently. If your Reels reference look is tied to a specific motion feel rather than a technical spec, test a clip on Kling before committing to a series.
Decision summary:
- Need a polished 9:16 hook or beat at 1080p with native audio → Veo 3.1 Standard
- Want 4 variants to pick the strongest take → Veo 3.1
- Need a single continuous clip longer than 8 seconds → Seedance 2.0 (4 to 15s)
- Need to hold a look across multiple takes → Seedance 2.0 omni-reference
- Need stylized motion outside the above → Kling 3.0
- Iterating cheaply before final render → Veo 3.1 Fast, then Standard for final
10 Reels Prompts, Copy, Label, and Generate
Each prompt below includes a recommended model. Swap subject, location, or product detail to fit your brand. Keep the structural vocabulary intact, it carries the vertical framing and aesthetic instruction load.
Prompt 1, Aesthetic Morning Routine Hook Recommended model: Veo 3.1 Standard (9:16, 1080p, 4 variants)
9:16 vertical, full-bleed subject, safe margins top and bottom. A woman in a linen robe opens curtains to soft morning light in a minimalist bedroom. Warm pastel palette, gentle bloom on the window highlight, soft film grain. Slow push in from medium to close-up. Soft window light from camera left, warm fill, no harsh shadow. Single clean beat, the light hits her face in the final second. Ambient room tone, no dialogue. 1080p.
Prompt 2, Coffee Shop Pour Reveal Recommended model: Veo 3.1 Standard (8s, 1080p, native ambient audio)
9:16 vertical, subject in upper two thirds. Top-down view of a barista pouring latte art into a ceramic cup on a wooden counter. Warm specialty-cafe aesthetic, soft daylight, shallow depth of field on the cup. Locked-off camera, only the pour moves. The leaf pattern completes in the final second as a clean sync beat. Ambient cafe audio, low chatter, espresso machine hiss, no dialogue. 1080p.
Prompt 3, Streetwear Outfit Turn Recommended model: Kling 3.0 (stylized motion, 9:16)
9:16 vertical, full-bleed subject, head to mid-thigh in frame. A young man in layered streetwear does a single slow turn against a concrete wall. Cool desaturated street palette, high contrast, crisp shadow. Hard practical light from camera right, cool rim from behind. Gentle handheld drift following the turn. Motion lands square to camera in the final second. No audio, clip scored in post. 9:16 vertical.
Prompt 4, Skincare Product Top-Down Recommended model: Veo 3.1 Standard (1080p, consistent descriptions for a product series)
9:16 vertical, product centered, safe margins for caption. Top-down on a glass serum bottle on a pale marble surface, a single water droplet falls and ripples beside it. Clean, premium, aspirational palette, soft diffused studio light, gentle highlight roll-off on the glass. Slow push in toward the bottle. The droplet lands in the final second. No audio, clip scored in post. 1080p.
Prompt 5, Golden-Hour Rooftop Lifestyle Recommended model: Veo 3.1 Standard (9:16, 1080p, ambient audio)
9:16 vertical, subject in upper two thirds with sky negative space above. A woman leans on a rooftop railing at golden hour, city skyline soft behind her. Warm aspirational palette, long low sun, gentle lens bloom on the backlight. Slow push in from wide to medium. Warm backlight, soft fill on her face from camera front. Hair moves slightly in the breeze. Ambient city air and distant traffic, no dialogue. 1080p.
Prompt 6, Food Plating Close-Up Recommended model: Veo 3.1 Standard (8s, 1080p)
9:16 vertical, dish centered, shallow margins. Close-up of a chef finishing a plate, a spoon drizzles sauce over a dessert on dark slate. Editorial food aesthetic, warm key light from camera left, deep falloff into shadow, glossy specular on the sauce. Locked-off camera, only the drizzle moves. The plate is complete on the final beat. Ambient kitchen tone, no dialogue. 1080p.
Prompt 7, Travel Walk-Through (Longer Take) Recommended model: Seedance 2.0 (12s, 9:16, native 9:16 supported)
9:16 vertical, full-bleed, subject centered with headroom. A traveler walks slowly through a narrow sunlit alley in an old town, hand trailing along a stone wall. Warm aspirational travel palette, dappled light through overhead foliage, soft film grain. Steady handheld follow from behind, then a gentle reveal of the square at the alley's end. Continuous 12-second take. Ambient footsteps and distant street sound, no dialogue. 9:16.
Prompt 8, Fitness Motion Beat Recommended model: Veo 3.1 Standard (9:16, 1080p, 4 variants)
9:16 vertical, subject head to knee in frame. A woman finishes a kettlebell swing in a sunlit home gym, motion freezing at the top of the rep. Clean, energetic palette, bright window light, crisp shadow, slight haze in the air. Locked-off camera, subject motion only. The rep peaks on the final beat for a clean sync point. Ambient gym tone, fabric and breath, no dialogue. 1080p.
Prompt 9, Cozy Interior B-Roll Recommended model: Seedance 2.0 (10s, 9:16, omni-reference for series consistency)
9:16 vertical, full-bleed interior. Slow drift across a cozy living room at dusk, a candle flickers on a wooden table, a throw blanket draped over a linen sofa, warm lamp glow in the background. Warm hygge palette, soft amber light, gentle film grain. Continuous 10-second slow lateral drift. The candle catches the lens in the final second. Ambient quiet room tone, no dialogue. 9:16.
Prompt 10, Unboxing Hook with Native Audio Recommended model: Veo 3.1 Standard (8s, 1080p, native ambient audio)
9:16 vertical, hands and product centered, safe margins. Close-up of hands opening a premium matte gift box on a clean desk, lid lifting to reveal a folded garment inside. Aspirational unboxing aesthetic, soft overhead studio light, gentle highlight on the box edge, shallow depth of field. Slow push in as the lid lifts. The reveal lands on the final beat. Ambient paper and fabric sound, no dialogue, no music. 1080p.
Five Mistakes That Produce Non-Reel-Ready Output
Most Reels prompts fail at the specification layer, not the creative concept. The clip reads generic because the prompt gave the model vague or contradictory instruction for the vertical frame.
Mistake 1: Writing for the wrong aspect ratio Prompting a wide scene and hoping to crop to 9:16 wastes the frame, the model composes for 16:9 and the subject ends up off-center or cut. Fix: state "9:16 vertical, full-bleed subject" in the first sentence so the model composes for the phone screen from the start.
Mistake 2: Ignoring caption and UI safe margins The Reels player covers the bottom strip with caption text, audio controls, and the action bar. A subject placed dead center or low gets obscured. Fix: keep the focal subject in the upper two thirds and state "safe margins for caption and UI."
Mistake 3: Packing multiple actions into a short clip "She walks in, sits down, opens a laptop, and waves" is four beats in 8 seconds, the model rushes all of them and none reads cleanly. Fix: one clip, one action, one clean beat the viewer can sync to a track.
Mistake 4: Saying "aesthetic" without naming the palette "Aesthetic Reel" gives the model no constraint, so it defaults to its most common output. Fix: name the palette and light (warm pastel with soft window light, cool desaturated street with hard rim) so the polished look is a specification, not a hope.
Mistake 5: Forgetting audio intent Veo 3.1 generates native audio. If you plan to add a trend track in post but do not say so, the model may layer ambient sound or dialogue that clashes. Fix: state "no audio, clip scored in post" for trend-track Reels, or name the ambient sound you want captured.
How to Iterate from Rough to Reel-Ready on muvi.video
A Reel rarely ships from the first generation. The workflow below reduces coin spend and compresses iteration by separating the rough composition pass from the style-lock pass.
Rough composition with Fast variant
Use Veo 3.1 Fast to test vertical framing, subject placement, and the single action beat. Skip palette and lighting nuance at this stage, confirm the 9:16 composition reads on a phone screen first. Fast produces output quickly and costs fewer coins than Standard.
Lock model and dial in the look with Standard variant
Once the composition works, switch to Veo 3.1 Standard and add the full vocabulary: named palette, lighting source, film grain, and the final-second sync beat. Because Veo 3.1 does not support deterministic seed control, use consistent prompt language across takes, and review the 4 variants it returns to pick the strongest hook. For tighter consistency across a series, use the Veo 3.1 Image-to-Video variant with a reference still.
Seedance 2.0 for clips longer than 8 seconds
For a continuous beat that needs to exceed 8 seconds, a walk-through or a long lateral drift, move to Seedance 2.0 after confirming the scene logic in a Veo 3.1 draft. Seedance supports any integer duration from 4 to 15 seconds and omni-reference to hold a look across takes. Resolution caps at 720p (1440p on the Start/End Frame variant), so reserve it for clips where duration matters more than pixel-level fidelity.
Final render and score in the Reels editor
Final delivery should use Standard variants at native 9:16. Do not crop a 16:9 output to vertical, it sacrifices resolution and reframes the composition. Export the clean clip, then add your trend track and captions in the Reels editor so the visual beat lands on the music drop.
Consistency across a series: Veo 3.1 does NOT support deterministic seed control, each run with the same prompt may vary. For a multi-clip Reel series, rely on consistent subject, wardrobe, and setting descriptions across prompts, anchor frames with the Veo 3.1 Image-to-Video variant, or use Seedance 2.0 omni-reference.
Instagram Reels AI Video Prompts, Frequently Asked Questions
Which muvi.video model is best for Instagram Reels?+
Veo 3.1 Standard at 1080p is the default for Reels on muvi.video, it composes natively for the 9:16 vertical frame, generates native ambient and dialogue audio, and returns 4 variants per run so you can pick the strongest hook. For a continuous beat longer than 8 seconds, Seedance 2.0 covers any integer duration from 4 to 15 seconds in native 9:16, with a lower resolution ceiling. For stylized motion, test Kling 3.0. Choose the model by duration and look first, then refine with the seven prompt components.
How do I make sure my AI Reel is the right 9:16 vertical size?+
State "9:16 vertical, full-bleed subject, safe margins for caption and UI" in the first sentence of your prompt. Veo 3.1 supports 9:16 natively, so it composes for the phone screen rather than cropping a wide frame. Keep your focal subject in the upper two thirds, the bottom strip is covered by the caption and audio controls in the Reels player. Generating native 9:16 always beats cropping a 16:9 clip, which loses resolution and reframes the shot.
How long can an AI-generated Reel clip be on muvi.video?+
Veo 3.1 generates clips of 4, 6, or 8 seconds at 1080p, which suits most single-beat Reels hooks. For one continuous clip longer than 8 seconds, Seedance 2.0 supports any integer duration from 4 to 15 seconds, at a lower resolution ceiling (720p, or 1440p on the Start/End Frame variant). For longer Reels, structure the edit as a series of short Veo 3.1 clips joined with your trend track rather than one long clip, which keeps each shot at full resolution.
Can I add trending music to an AI-generated Reel?+
Generate the visual clip on muvi.video, then add your trend track and captions in the Instagram Reels editor. When you prompt, state "no audio, clip scored in post" so Veo 3.1 does not layer ambient sound that clashes with your track, and build a clean visual beat (a reveal, a turn, a pour) that lands in the final second so it syncs to a music drop. If you want captured ambient sound instead, name it in the prompt, for example "ambient cafe audio, no dialogue."
How do I keep a consistent look across a Reels series?+
Veo 3.1 does not support deterministic seed control, so for a consistent series rely on a named palette and consistent subject, wardrobe, and setting descriptions in each prompt, and anchor frames with the Veo 3.1 Image-to-Video variant by feeding a reference still from a prior clip into the next. Seedance 2.0 also offers omni-reference to carry a look across takes. Lock the palette and lighting language once, then reuse it verbatim across every prompt in the series.
How is Veo 3.1 billed across plans?+
Veo 3.1 generations are coin-based across all plans, and the included monthly coin allotment differs by tier. The Ultra Yearly plan includes unlimited Veo 3.1. For current per-generation coin costs and monthly allotments by plan, see the pricing page.
Why do my AI Reels look generic instead of aesthetic?+
The usual cause is vague style language. "Aesthetic Reel" gives the model no constraint, so it defaults to its most common output. Name the palette by reference (warm pastel, cool desaturated street, hygge amber), describe the light source and quality (soft window light, hard practical neon), and define one clean action beat. Narrow the instruction space and the polished, aspirational look becomes a specification rather than a hope.
Generate Instagram Reels AI Video on muvi.video
Apply these prompts in muvi.video Studio. Select Veo 3.1 Standard for polished 9:16 hooks and beats at 1080p with native audio and 4 variants to choose from. For continuous clips longer than 8 seconds, switch to Seedance 2.0. Move from first draft to Reel-ready vertical video in a single workspace, then add your trend track in the editor.
No credit card required · Works in your browser · Veo 3.1, Seedance 2.0, and Kling 3.0 available
More Resources
Instagram Reels AI Video Prompts: 10 Copy-Ready Prompts (2026)
Ten copy-ready Instagram Reels AI video prompts for 9:16 vertical on muvi.video, built for Veo 3.1, Seedance 2.0, and Kling 3.0 with aesthetic, trend-aware framing.