Text to video and image to video

AI Video Prompts for Veo 3, Sora, Kling and 6 More Models

One shot, one camera move, one clear action. The studio turns your idea into a video prompt in the format each model reads: Veo 3 dialogue and SFX lines, Sora action beats, Hailuo camera commands, Higgsfield presets.

Build a prompt See examples

Free, no sign-up

15 models, image and video

Updated October 2026

AI model What do you want?

Describe it Keep exactly the same

Setting Style

Lighting Shot / framing

Camera move Lens look Format

Exact text in the image (optional)

Sound (optional) Spoken line (optional)

Avoid (optional) Stylize --s (0-1000, optional)

Your prompt for Veo 3

Describe what you want on the left. Your prompt appears here as you type.

Open Veo 3

Free: 5 AI runs a day, 3 variations each. Pro: 100 runs, 10 variations, saved prompt library.

Tips

Keep each spoken line short enough to say in a few seconds; one speaker per shot lip-syncs best.
Set aspect ratio and clip length in Flow or the Gemini app settings.

Image to prompt

Upload a picture and the AI writes a prompt that recreates it for the model picked in the studio above, in that model's format. Your image is processed in memory and never stored.

Your image (JPEG, PNG or WebP)We shrink it to 1024 px in your browser before sending. It is not saved on our side. Note for the AI (optional)

Pro feature. Image to prompt is part of All-Access Pro: 20 images a day, plus 100 AI runs, 10 variations and a saved prompt library. Already Pro? Sign in.

Prompt from your image for the selected model

Pick a model in the studio above, upload an image and press "Write prompt from image".

AI Video Prompts: examples you can copy

Each example was written by the studio above. Click "Edit in the studio" to load it into the form and change the details.

Veo 3 talking headVeo 3

Medium close-up of a presenter in her 30s with a short black bob and a mustard sweater looks straight into the camera, in a bright home studio. Camera: locked-off static camera. Lighting: soft key light, natural skin. Dialogue: the speaker says in a clear, natural voice: "Three prompts that saved me ten hours this week." No subtitles. SFX: clean voice, light room tone, no music.

Sora product heroSora 2

Animate the attached image. Keep the bottle shape, label text, logo and colors exactly as in the image. Cinematography: Camera motion: smooth half orbit around the subject, keeping it centered and in focus Lighting: rim light, dark background Actions: - A soft light sweep moves across the bottle from left to right - Condensation drops slide down the glass

Kling action shotKling

Wide tracking shot of a skateboarder in a red hoodie rolls along a seaside promenade and kicks into an ollie. Camera: the camera trucks sideways at walking speed, parallel to the subject. Lighting: hard late-afternoon sun, long shadows. Style: sports commercial, high contrast.

Negative prompt: camera shake, morphing, face distortion, extra limbs, flicker, text, watermark

Runway portrait comes aliveRunway

The woman blinks naturally, takes a breath and slowly turns her head toward the camera with a soft smile. Slow, steady dolly in toward the subject. Cinematic, soft natural light.

Hailuo revealHailuo (MiniMax)

[Pedestal up, Push out] Close-up of hands plant a seedling in a rooftop garden as the city wakes up. Lighting: soft morning light. Style: cinematic, natural colors, fine grain.

Higgsfield product zoomHiggsfield

Animate the attached image: the sneaker stands on wet asphalt; the move starts wide and ends on a tight close-up of the logo. Keep the sneaker shape, colors and logo exactly as in the image.

What each video model expects

All of them read subject, action, setting, camera, light, style and sound. What differs is the syntax around it.

Veo 3

Generates sound with the picture. Dialogue in quotes after the named speaker, SFX and ambient noise on their own lines, "no subtitles".

Sora 2

A short scene description, then a cinematography block (shot, camera motion, lighting) and actions as small beats.

Kling and Pika

Prose prompt plus a separate negative prompt field for camera shake, morphing and flicker.

Runway

Gen-4 animates an image: describe the motion only, phrase everything positively, no negatives.

Higgsfield

Pick a camera preset in the app (Dolly In, Crane Up, FPV Drone); the text must not describe another move.

Hailuo, Seedance, Luma

Hailuo reads [Push in] bracket commands; Seedance likes a production brief with audio; Luma wants subject, action, scene, style, then camera.

How to write an AI video prompt

One shot, one action with a start and an end "lifts the cup, takes a sip, looks out of the window". Three actions is the practical ceiling for one clip.
One camera move with a speed "slow, steady dolly in" beats "dynamic epic camera". Two moves in one sentence is the top cause of warped clips.
Image to video: motion only The image already sets subject and light. Describe what moves and what must stay exactly the same.
Plan the sound Ambience, effects, music and one short spoken line, only for models and modes that generate audio.
Set ratio and length in the tool Settings win over words. 9:16 for Reels and TikTok, 16:9 for YouTube.

AI video models: prompt cheat sheet

Checked October 2026. Model syntax changes often; we review this table every month.

Model	Modes	Negative prompt	Sound from prompt	Key tip
Veo 3 Google (Gemini app, Flow)	Video from text, Video from an image	One Avoid sentence	Yes (check your mode)	Keep each spoken line short enough to say in a few seconds; one speaker per shot lip-syncs best.
Sora 2 OpenAI	Video from text, Video from an image	One Avoid sentence	Yes (check your mode)	Describe actions in beats ("takes two steps, stops, looks up"), not vague moods.
Kling Kuaishou	Video from text, Video from an image	Separate field	Yes (check your mode)	Paste the negative prompt into Kling's negative prompt field.
Runway Runway (Gen-4, Gen-4.5)	Video from an image, Video from text	None, phrase positively	No	Build it in steps: run the first sentence alone, then add camera, then scene motion, then style.
Higgsfield Higgsfield (DoP, camera presets)	Video from an image, Video from text	One Avoid sentence	No	Pick the camera preset in the app; the text must not describe another camera move.
Hailuo (MiniMax) MiniMax	Video from text, Video from an image	One Avoid sentence	No	Turn the prompt optimizer off when your exact wording must survive.
Seedance ByteDance	Video from text, Video from an image	One Avoid sentence	Yes (check your mode)	Say what each reference image is for (character, product, background).
Luma Ray Luma Dream Machine	Video from text, Video from an image	None, phrase positively	No	Use keyframes (start / end frame) for transitions and the loop option for banners.
Pika Pika	Video from text, Video from an image	Separate field	No	Put the negative prompt into Pika's negative prompt field.

Want ready prompts instead of building them?

The AI Video Prompt Pack has 54 tested prompts with formulas, a camera and style cheat sheet, troubleshooting and an assistant prompt for Claude or ChatGPT.

AI Video Prompt Pack

54 prompts as a plain-text file, yours to keep. One-time payment.

Get the pack, $4

Or everything with Pro

100 AI runs a day with 10 variations, prompts from your images, a saved library of 500 prompts with tags and search, every prompt pack, and 10 files a month.

See Pro

More prompt tools

AI image prompts Nano Banana prompts Midjourney prompt generator ChatGPT image prompts ChatGPT photo editing prompts Gemini prompts DALL-E prompts Prompt optimizer for text AI

FAQ

What is a good AI video prompt?

Shot size, a subject with 2-3 details, one action with a start and an end, the setting, one camera move with speed, light, style and, for models with sound, an audio line.

Which AI video models does it support?

Veo 3, Sora 2, Kling, Seedance, Runway, Higgsfield, Hailuo (MiniMax), Luma Ray and Pika. Each gets its own format.

Do Veo 3 prompts need a special format for dialogue?

Name the speaker, put the exact words in double quotes, describe the voice and add "no subtitles". Short lines lip-sync best.

Why does my AI video warp or the camera jitter?

Usually two camera moves in one prompt or vague words like "dynamic". Use one move from the list, add "slowly", and lock the camera off for labels and faces.

Are Sora prompts different from Kling prompts?

Yes. Sora likes a cinematography block and action beats; Kling takes one prose paragraph and a negative prompt field. Pick the model and the studio switches format.

Is the AI video prompt generator free?

Yes. Prompts are built in your browser with no limit. AI variations: 5 runs a day free, 100 with Pro, 10 variations at a time.

Can it write a prompt from my image?

Yes, with Pro. In Image to prompt upload a picture, pick the model in the studio, and the AI writes a prompt that recreates it in that model's format, up to 20 images a day. The image is shrunk in your browser, processed in memory and never stored by us; press Edit in the studio to tweak the result.