KIVIDE AIGenerate AI videos & images with premium models. No credit card needed.
Type your idea in plain English, pick a target model, and get a structured prompt split into shot, subject, action, lighting, camera, style and quality. Free, no signup required.
Build a video prompt
Target model
Write it like a shot brief with an audio cue; camera and sound are both first-class.
Your structured prompt
VeoModel dialects
Each model was trained on a different kind of caption and listens for different words. Veo treats camera and sound as first-class, so naming the lens and an audio line changes the result far more than adjectives do. Kling rewards one clearly described physical action and honours negations, but stacks of simultaneous actions turn rubbery. Seedance expects ordered beats and will collapse a vague timeline into a single shot. Sora reads narrative prose and mostly ignores negative prompts, so you steer it by describing what should happen, not what should not. Runway is camera-first and punishes long literary sentences that bury the move. Wan is literal: concrete nouns and motions work, metaphor does not. Writing one generic caption and pasting it everywhere is the single most common reason a good idea renders poorly.
How the six mainstream models want to be written to. Treat this as use-case guidance, not a quality ranking — these models change often.
| Model | Best at | Prompt length | Negative prompts | Common pitfall |
|---|---|---|---|---|
| Veo | Cinematic realism, native audio | Medium–long (30–60 words), with an explicit audio line | Sometimes | Describe sound explicitly and add “(no subtitles)” or the model may render caption text. |
| Kling | Character motion, physical realism | Medium (20–40 words), one clear action | Supported | Too many simultaneous actions make the motion rubbery — keep to one primary movement. |
| Seedance | Multi-shot sequencing, timelines | Long (40–80 words), ordered beats | Supported | Vague sequencing makes the timeline collapse into one shot — order the beats explicitly. |
| Sora | Complex scenes, narrative | Long (50–100 words), prose | Not reliable | List-like fragments read as flat — write full sentences with a sense of story. |
| Runway | Image-to-video, camera control | Short (10–25 words), camera-first | Sometimes | Long literary prompts dilute the camera instruction; keep it tight and motion-led. |
| Wan | Cost-efficient clips, product demos | Medium (20–40 words), concrete clauses | Sometimes | Abstract or poetic language is ignored — name concrete objects and motions. |
A cross-section of the most common briefs, each tagged with the model it suits. Every prompt below is also plain text in this page's HTML, so you can copy it with JavaScript disabled.
a lone astronaut walking across a red desert at dusk
Shot: Single shot, 16:9, ~5 seconds Subject: A lone astronaut walking across a red desert at dusk Action: A lone astronaut walking across a red desert at dusk Lighting: Warm golden-hour light, long soft shadows Camera: Slow dolly-in, 35mm lens, shallow depth of field Style: Photoreal, cinematic grade, natural motion Quality: Sharp focus, natural film grain, no text or watermarks
a glass perfume bottle rotating on a black surface
Shot: Single shot, 16:9, ~5 seconds Subject: A glass perfume bottle rotating on a black surface Action: A glass perfume bottle rotating on a black surface Lighting: Soft directional key light, balanced fill Camera: Camera pushes in from a low angle Style: Clean commercial look, controlled motion Quality: Sharp focus, natural film grain, no text or watermarks
a founder explaining their product in a bright office
Shot: Single shot, 16:9, ~5 seconds Subject: A founder explaining their product in a bright office Action: A founder explaining their product in a bright office Lighting: Soft directional key light, balanced fill Camera: Medium tracking shot, eye-level, natural handheld weight Style: Realistic, physically believable motion Quality: Sharp focus, natural film grain, no text or watermarks
a young hero drawing a glowing sword in the rain
Shot: Two-shot sequence, 16:9, ~8 seconds total Subject: A young hero drawing a glowing sword in the rain Action: A young hero drawing a glowing sword in the rain Lighting: Soft directional key light, balanced fill Camera: Shot 1: wide establishing; Shot 2: medium close-up Style: Sequenced, continuity across shots Quality: Sharp focus, natural film grain, no text or watermarks
a neon street at night with a retro sports car
Shot: Single shot, 16:9, ~5 seconds Subject: A neon street at night with a retro sports car Action: A neon street at night with a retro sports car Lighting: Low-key lighting, practical sources, deep shadows Camera: Static medium shot with a gentle pan right Style: Clean, efficient, product-demo clarity Quality: Sharp focus, natural film grain, no text or watermarks
an electric car speeding along a coastal highway
Shot: Single shot, 16:9, ~5 seconds Subject: An electric car speeding along a coastal highway Action: An electric car speeding along a coastal highway Lighting: Soft directional key light, balanced fill Camera: Slow dolly-in, 35mm lens, shallow depth of field Style: Photoreal, cinematic grade, natural motion Quality: Sharp focus, natural film grain, no text or watermarks
a waterfall in a misty forest at sunrise
Shot: Single continuous moment, 16:9, ~10 seconds Subject: A waterfall in a misty forest at sunrise Action: A waterfall in a misty forest at sunrise Lighting: Warm golden-hour light, long soft shadows Camera: Smooth crane rise into a wide two-shot Style: Narrative, filmic, emotionally grounded Quality: Sharp focus, natural film grain, no text or watermarks
a spaceship emerging from a nebula
Shot: Single continuous moment, 16:9, ~10 seconds Subject: A spaceship emerging from a nebula Action: A spaceship emerging from a nebula Lighting: Soft directional key light, balanced fill Camera: Smooth crane rise into a wide two-shot Style: Narrative, filmic, emotionally grounded Quality: Sharp focus, natural film grain, no text or watermarks
a person unboxing headphones at a desk, phone camera style
Shot: Single shot, 16:9, ~5 seconds Subject: A person unboxing headphones at a desk, phone camera style Action: A person unboxing headphones at a desk, phone camera style Lighting: Soft directional key light, balanced fill Camera: Camera pushes in from a low angle Style: Clean commercial look, controlled motion Quality: Sharp focus, natural film grain, no text or watermarks
steam rising off a bowl of ramen in a warm kitchen
Shot: Single shot, 16:9, ~5 seconds Subject: Steam rising off a bowl of ramen in a warm kitchen Action: Steam rising off a bowl of ramen in a warm kitchen Lighting: Soft directional key light, balanced fill Camera: Medium tracking shot, eye-level, natural handheld weight Style: Realistic, physically believable motion Quality: Sharp focus, natural film grain, no text or watermarks
a golden retriever running on a beach
Shot: Two-shot sequence, 16:9, ~8 seconds total Subject: A golden retriever running on a beach Action: A golden retriever running on a beach Lighting: Warm golden-hour light, long soft shadows Camera: Shot 1: wide establishing; Shot 2: medium close-up Style: Sequenced, continuity across shots Quality: Sharp focus, natural film grain, no text or watermarks
a camera gliding through a modern glass house
Shot: Single shot, 16:9, ~5 seconds Subject: A camera gliding through a modern glass house Action: A camera gliding through a modern glass house Lighting: Soft directional key light, balanced fill Camera: Static medium shot with a gentle pan right Style: Clean, efficient, product-demo clarity Quality: Sharp focus, natural film grain, no text or watermarks
Name who or what is on screen and one identifying detail — “a lone astronaut in a scuffed white suit” beats “someone lonely”. Vague subjects make the model invent details you then have to fight. One subject per prompt keeps focus.
Say the movement and the lens: dolly-in, crane rise, handheld tracking, 35mm, shallow depth of field. Camera-first models such as Runway need this in the first clause; putting it last buries it. Avoid stacking two moves in one shot.
Describe the light source, not the emotion: “warm golden-hour key light, long shadows” or “low-key neon from the side”. Lighting drives realism more than any other field and pairs directly with the time of day in your idea.
Kling and Seedance respond well to explicit avoid-lists (extra limbs, warped text, duplicates). Sora largely ignores negations, so with Sora describe the correct outcome instead. When in doubt, rephrase “no blur” as “sharp, in focus”.
Yes. You can build, copy and regenerate prompts without signing up. Generating the actual video needs an account, and new accounts get free credits on signup.
The builder covers six mainstream families — Veo, Kling, Seedance, Sora, Runway and Wan — each with its own prompt dialect, length guidance and negation behaviour. Kivide's studio runs the production engines behind them, and you can switch models there without rewriting your prompt.
Yes, and that is the point of this page. The same idea written for Runway (short, camera-first) will underperform on Seedance (ordered beats) and vice versa. Pick your target model above and the structure adjusts.
Only where the model honours them. Kling and Seedance react to explicit avoid-lists; Sora mostly ignores them and is steered by describing the desired result instead. The cheat sheet above shows which is which.
Short for camera-control models such as Runway (10–25 words), longer for sequencing models such as Seedance (40–80 words) and narrative models such as Sora (50–100 words). The recommended length for each model is listed in the table.
