Skip to content
promptintermediateFree

Video prompt translator (Veo, Sora, Kling, Runway, Luma, Hailuo)

Rewrite one shot idea into the phrasing style each video model tends to respond to, so you can test the same shot across tools fairly.

You'll need: A chat model · A shot idea or an existing prompt

Fill in your details

Values are inserted into every prompt below. Nothing you type leaves your browser.

5 field(s) left — unfilled variables stay as {{placeholders}}.

  1. 1Translate the prompt

    Run in a chat model. The style notes are general tendencies, not official rules. Check each tool's current prompt guide, since models update often.

    Rewrite this video shot for several video generation models. Keep the same subject, action, setting, camera, lighting, mood, duration and aspect ratio in every version.
    
    Shot: {{shot}}
    Duration: {{duration}}. Aspect ratio: {{aspect_ratio}}.
    Dialogue or sound (optional): {{audio}}
    Target models: {{target_models}}
    
    Apply these general phrasing tendencies (adjust if you know the tool's current guidance better):
    - Detailed natural-language models (e.g. Veo, Sora): flowing descriptive paragraph, cinematography terms, sound and dialogue described in their own sentences if the model supports audio.
    - Concise camera-first models (e.g. Runway, Luma): short prompt, camera movement first, then subject and action, minimal adjectives.
    - Structured prompt models (e.g. Kling, Hailuo, Wan, Seedance): clear order of subject, action, setting, camera, lighting, style; put unwanted elements in the negative field if the tool has one.
    
    Output one labelled prompt per target model, then a 2-line note on what to compare between the results.

Example input

Lighthouse keeper climbing a spiral staircase with a lantern in a storm; 6s; 16:9; storm sounds.

Example output

Veo-style: A low-angle shot follows a weathered lighthouse keeper in a dark oilskin coat as he climbs an iron spiral staircase, a swinging lantern throwing warm moving shadows across the curved stone walls. Storm light flashes through a narrow window. Moody, cinematic, 6 seconds, 16:9. Audio: wind howling, rain lashing the glass, heavy footsteps ringing on iron steps. Runway-style: Low-angle tracking shot moving up. Lighthouse keeper climbs a spiral iron staircase holding a lantern. Storm flashes through a window. Warm lantern light, dark stone. Cinematic. Kling-style: Subject: weathered lighthouse keeper in a dark oilskin coat holding a lantern. Action: climbs an iron spiral staircase. Setting: stone lighthouse interior during a storm. Camera: low angle, following upward. Lighting: warm lantern light, cold lightning flashes. Style: cinematic, moody. Negative: text, watermark, distorted hands, extra people.

Troubleshooting

  • A model ignores the camera: move the camera phrase to the start for that model.
  • Compare fairly: use the same duration, aspect ratio and similar quality settings for each tool.

Known limitations

  • Model behaviour changes with each release; these style tendencies are starting points, not documented rules.
  • Audio lines only matter for models and plans that generate sound.

More from Video generation