DEV Community: OpenVideos AI

How I structure 9:16 AI short prompts with a timed beat sheet

OpenVideos AI — Wed, 29 Jul 2026 09:37:34 +0000

Originally published at openvideos.ai.

If you generate a 16:9 clip and crop it for TikTok / Reels / Shorts, the model composed for a wide stage. Heads get clipped and the payoff lands in the wrong third of a phone screen.

I write vertical shorts as a timed beat sheet instead: one subject, one location, and costume/camera changes on a clock.

The skeleton

[Title / vibe]
0s–Xs, [Shot size], [Look A + one action + light]
Xs–Ys, [Camera move], [Transition / look B]
Ys–Zs, [Pull back], [Finale pose + light accent]

Rules that matter in 9:16:

One identity across beats (the tall frame magnifies face drift)
Explicit shot sizes: medium → close-up → medium
One camera verb per beat
Name the second when the outfit/light changes

A verified 10s run

On OpenVideos I ran Seedance 2.0 (seedance-2-t2v) as text-to-video:

Setting	Value
Duration	10s
Aspect	`9:16`
Resolution	`480p`
Audio	on
Credits (registry)	520 (= 10 × 52 at 480p)

Prompt (abridged structure — full text on the canonical post):

0–3s medium / eye level — plain ruqun, warm light, small hem lift
3–7s push to close-up — tear/splash into cheongsam, hard contrast
7–10s pull back — embroidered battle look, cold metal accents, pose that loops

That “look A → transition hit → finale pose” template also remaps to streetwear, beauty, or product unbox. Keep the clock; swap the world.

Compose in the prompt, not only the picker

Setting aspect_ratio: 9:16 is necessary but not enough. Prefer medium/close-up, put the face in the upper third, leave a quiet band for captions, and favor vertical motion over long lateral pans.

Full annotated prompt, failure table, and the playable result live on the original guide:

→ AI Short Video Prompts for 9:16

How to Write AI Video Prompts Without Rewriting Everything

OpenVideos AI — Wed, 29 Jul 2026 05:20:38 +0000

Originally published at openvideos.ai.

Start from the input you already have

The useful prompt formula changes with what you bring in.

Text-to-video: the prompt must define the whole shot: subject, action, place, framing, camera, light, and timing.
Image-to-video: the image already carries subject, composition, light, and style. The prompt should mainly describe what moves.

Repeating a long text-to-video description on top of a finished still is a common failure mode. The model gets several competing versions of the same scene.

Formula 1: text-to-video as a seven-part shot brief

Subject + Action + Setting + Shot + Camera movement + Light/style + Timing

You do not need a long sentence for every part. You need enough information to remove the decisions that matter.

Part	Question	Example
Subject	What must stay recognizable?	A lone cyclist in a mustard-yellow rain jacket
Action	What changes in this shot?	Rides steadily across the bridge
Setting	Where and when?	Blue hour after rain, wet road, light mist
Shot	What belongs in the frame?	Wide road-level three-quarter shot
Camera	How does the viewpoint change?	Tracks parallel at a steady distance
Light/style	What is physically visible?	Cool ambience, warm road reflections
Timing	How should motion unfold?	One controlled, continuous shot

Keep short clips focused on one scene. If the idea contains three plot beats, split them into separate shots.

A complete example

A lone cyclist in a mustard-yellow rain jacket rides steadily across an empty suspension bridge at blue hour after rain. Wide road-level three-quarter shot, with the cyclist on the left third and open road ahead. The camera tracks parallel at a steady distance. Wet asphalt reflects distant red and amber traffic lights; cool overcast ambience, light mist, restrained contemporary film color. One continuous shot with controlled speed and smooth tracking throughout.

The prompt is detailed because it resolves seven decisions, not because more words automatically produce a better video.

Formula 2: image-to-video should describe motion

Camera motion + Subject motion + Environmental motion + Timing

Treat the source image as the visual foundation. For the cyclist still, a motion prompt can stay short:

The camera tracks smoothly beside the cyclist. The cyclist pedals at a steady pace while the rain jacket moves lightly in the wind. Fine mist drifts across the bridge and road reflections shimmer subtly. Controlled motion, one continuous shot.

Skip a second biography of face, jacket color, bridge architecture, or lighting. The image already supplies those facts.

Diagnose before you rewrite

After a weak result, name the smallest failed decision first.

What you observe	Likely problem	First revision
Subject looks generic	Missing visible anchors	Add two or three defining details
Too much happens	Competing events	Keep one primary action
Framing feels accidental	No shot size or viewpoint	Add one explicit shot instruction
Motion is confused	Camera and subject fight	Keep one camera move and one subject action
Image barely moves	Prompt re-describes the still	Remove scene description; state motion
Mood is vague	Abstract adjectives	Specify light, color, weather, or texture

Change one component at a time. Keep the rest stable so the next generation answers a real question.

A credit-aware loop

Choose the input mode before polishing language.
Write the minimum complete prompt for one shot.
Inspect the selected model's actual controls.
Confirm displayed credits before submit.
Save the exact prompt and result.
Name the largest mismatch.
Change one component.

Sources and limits

This framework synthesizes first-party guidance from Google, Runway, and Luma video-prompt docs. It does not rank models or claim identical results across them.

Full guide with checklist and links: How to Write AI Video Prompts.

How to Make Cinematic AI Video (A Practical Decision Guide)

OpenVideos AI — Tue, 28 Jul 2026 07:21:53 +0000

Who this is for

This guide is for creators who want cinematic AI video—controlled framing, lighting, and camera language—not random “AI music video” motion. You should already know you will generate with a credit-based studio (OpenVideos), not an unlimited free sandbox.

After reading, you should be able to:

Decide whether your shot is text-to-video, image-to-video (start frame), or first/last frame.
Write a prompt that locks subject + camera + light + motion + continuity.
Open OpenVideos Create with the right category and confirm credits before submit.

This page is informational. The transactional cinematic spoke lives at /ai-video-generator/cinematic. Use this guide to plan; use the spoke or Create when you are ready to spend credits.

What “cinematic” means here (and what it does not)

In practice, cinematic AI video usually means four controllable ingredients:

Ingredient	What to specify	Common failure
Framing	lens feel, subject size, angle	subject floats; no eyeline
Light	key/fill/rim, time of day, contrast	flat “AI default” lighting
Motion	camera move + subject move, speed	jitter, morphing, teleport
Continuity	wardrobe, location, prop identity	face/outfit drift mid-clip

It does not mean “add the word cinematic to any prompt.” That word alone rarely fixes composition.

Decide the workflow before you write poetry

OpenVideos Create exposes distinct video categories:

text-to-video — prompt-only. Best when you do not have a locked hero still.
start-frame-video — you supply a first frame (first_frame_image). Best when composition is already approved as a still.
start-end-frame-video — first + last frame (last_frame_image supported on capable models). Best when you need a planned start and end pose/composition.
ref-to-video / video-editor — later-stage refinement; not the default first cinematic attempt.

Rule of thumb

If you are still exploring look and location → start with text-to-video.
If you already love a still (generated or photographed) → start-frame-video.
If the story is “from pose A to pose B” → start-end-frame-video (see the upcoming workflow post on first/last frame).

Product path shortcuts:

Cinematic spoke (examples + Create entry): /ai-video-generator/cinematic
Text-to-video hub: /text-to-video
Image-to-video hub: /image-to-video
Pricing / credit packs: /pricing

A prompt formula that survives motion

Use this stack (order matters because models overweight early tokens):

Subject identity — age range, wardrobe, distinctive props (one sentence).
Environment — place, weather, time of day, depth cues.
Composition — shot size (close-up / medium / wide), angle, alone vs group.
Lighting — key direction + quality (soft/hard) + practical sources if any.
Camera move — one primary move (slow push-in, lateral track, locked-off).
Subject action — one primary action; avoid stacking five verbs.
Negative constraints — no text, no watermark, no morphing hands (short list).

Copy-ready example (text-to-video)

Photoreal medium close-up of a woman in her late 20s with damp chestnut hair and a black leather jacket, standing alone on a rainy city street at night. Neon cyan/magenta reflections on wet asphalt. Slight low angle, shallow depth of field. Soft rain streaks, cool neon rim from camera-left, warm shop-window fill on her face. Camera: slow push-in over 4 seconds, stabilized. Subject action: she turns toward camera and holds eye contact; minimal head movement. Natural pores, filmic grain, cinematic color grade. No text, no watermark, no logos, no readable neon lettering.

Why this works better than “cinematic rainy girl walking forever”:

One subject action + one camera move.
Light directions are explicit.
Continuity risks (readable neon, logos) are constrained.

Production checklist (before you spend credits)

One job, one intent — do not mix “epic drone establishing + extreme close-up dialogue” in a single short clip.
Lock aspect ratio for the destination (for example 9:16 shorts vs wider editorial). Confirm the ratio in Create before submit.
Estimate credits in Create — OpenVideos is pay-per-generation. The create panel shows the cost for the selected model, duration, and options. Do not assume a flat “one credit = one video.”
Know your balance — new accounts receive 50 signup bonus credits valid 30 days. Packs on /pricing are one-time purchases with their own validity windows; nearest-expiry credits are consumed first.
Save the still if you iterate — when a text-to-video look is right but motion fails, regenerate a still (image) and switch to start-frame-video instead of rewriting the whole story.

Common failures and fixes

Symptom	Likely cause	Fix
Face morphs mid-clip	Too much action + weak identity	Shorter action; start from a still
“Soap opera” motion	Competing camera + subject moves	Keep one move dominant
Flat look	No light direction	Add key/rim + practical source
Busy neon soup	Unconstrained signage	Ban readable text/logos
Great still, bad video	Wrong category	Move to start-frame / first-last frame

Limits (read before you publish claims elsewhere)

This guide does not claim which model is “best” or fastest. Model rankings need a controlled benchmark (same prompt, logged credits, retries). Those belong in dedicated benchmark posts—not here.
Exact per-job credit cost depends on the model and parameters shown in Create at submit time; it is not hard-coded in this article.
OpenVideos generation requires account credits; signup bonus is limited and time-bounded—not unlimited free generation.

Weak next step

When you have a single shot brief, open Create on text-to-video, paste a formula-built prompt, confirm credits, and generate one short clip. If you already have a hero still, prefer the image-to-video path instead.

Related product surfaces: /ai-video-generator/cinematic · /text-to-video · /pricing.

I stopped buying five AI subscriptions — one credit wallet for image + video

OpenVideos AI — Thu, 23 Jul 2026 06:30:30 +0000

Last year my browser toolbar looked like a SaaS graveyard.

One tab for image gen. Another for image-to-video. A third for "cinematic" experiments. A fourth because the first three were rate-limited on the weekend I actually needed them.

I was not building a production pipeline. I was managing five logins and five invoices.

The real cost was not money

Yes, stacked subscriptions add up. The worse cost was context switching:

Prompt that worked in tool A failed in tool B
"Download → re-upload" became the entire workflow
I could never answer: how much did this one clip actually cost?

If you are shipping creatives weekly, opacity kills iteration speed.

What I wanted instead

A single Create surface where I can:

pick a model per job (image or video)
spend from one credit wallet
keep outputs in one place

That is why I built OpenVideos.ai — a credit-pack studio for AI video + images, not another monthly seat I forget to cancel.

New accounts get 50 starter credits. No "unlimited forever" fairy tale. You see spend as you generate.

What still hurts (honest)

Credit math is not intuitive until you generate a few times — we show estimates, but you learn by doing
Multi-model does not mean "every model on earth on day one"
If you need a full editor suite (timeline, color grade, team seats), this is not Final Cut

The workflow that stuck

Steal a look from /inspiration
Open Create with a prompt + model
Iterate until the still works
Push the still into video when the shot is worth it
Download from My Works

If you are still copy-pasting between five AI tabs, try collapsing to one wallet for a week. Worst case you spend a few credits. Best case you get your evenings back.

→ openvideos.ai

I stopped buying five AI subscriptions — one credit wallet for image + video

OpenVideos AI — Thu, 23 Jul 2026 05:25:17 +0000

Last year my browser toolbar looked like a SaaS graveyard.

One tab for image gen. Another for image-to-video. A third for "cinematic" experiments. A fourth because the first three were rate-limited on the weekend I actually needed them.

I was not building a production pipeline. I was managing five logins and five invoices.

The real cost was not money

Yes, stacked subscriptions add up. The worse cost was context switching:

Prompt that worked in tool A failed in tool B
"Download → re-upload" became the entire workflow
I could never answer: how much did this one clip actually cost?

If you are shipping creatives weekly, opacity kills iteration speed.

What I wanted instead

A single Create surface where I can:

pick a model per job (image or video)
spend from one credit wallet
keep outputs in one place

That is why I built OpenVideos.ai — a credit-pack studio for AI video + images, not another monthly seat I forget to cancel.

New accounts get 50 starter credits. No "unlimited forever" fairy tale. You see spend as you generate.

What still hurts (honest)

Credit math is not intuitive until you generate a few times — we show estimates, but you learn by doing
Multi-model does not mean "every model on earth on day one"
If you need a full editor suite (timeline, color grade, team seats), this is not Final Cut

The workflow that stuck

Steal a look from /inspiration
Open Create with a prompt + model
Iterate until the still works
Push the still into video when the shot is worth it
Download from My Works

If you are still copy-pasting between five AI tabs, try collapsing to one wallet for a week. Worst case you spend a few credits. Best case you get your evenings back.

→ openvideos.ai

I stopped juggling five AI tools for video and images

OpenVideos AI — Wed, 22 Jul 2026 15:08:57 +0000

Most weeks I was bouncing between one tab for images, another for video, and a third just to remember which account still had credits.

That friction is small each time — and expensive across a month of shipping demos, social clips, and product mockups.

What I wanted

One place to generate images and video
The option to switch models without migrating projects
Credits I can buy when I need them, not another seat I forget to cancel

What I built around instead

I started using OpenVideos.ai as a single Create workspace:

Open Create
Pick a model for that job
Generate (image or video)
Pull results from My Works

For cinematic / short-form experiments I usually start from the video hub:

➡️ AI video generator

If I need reference samples or prompt ideas first:

➡️ Inspiration gallery

Honest notes

It's credit-based (pay-as-you-go packs), not "unlimited forever"
New accounts get 50 starter credits to try the flow
I'm not claiming it replaces every desktop suite — it's for shipping visuals without tool-hopping

If you're in the same boat

Try the Create panel once with a real prompt you'd actually use this week. If the multi-model + credits model fits, keep going. If not, at least you'll know after one session instead of five tabs.

Link: https://openvideos.ai

Happy to hear what model-switching friction still bothers you most.