Skip to content

Prompt Tips

The quality of your AI-generated output depends on how you write your prompt. This is a quick-start — for the comprehensive framework with camera language and shot recipes, see Prompting for Video Editors in Academy.

modelBridge includes a one-click prompt optimizer. Click the sparkle button below the prompt field, and an AI rewrites your description into a model-optimized prompt.

  • Cost: approximately $0.01 per use (shown in hover tooltip)
  • How it works: your prompt is sent to an LLM that rewrites it with better structure and specificity
  • The button is disabled when the prompt field is empty
  • Undo: the optimized prompt replaces your text — Cmd+Z to revert

Write prompts like shot descriptions: subject + action + environment + style + camera. 15–40 words. Specific and short beats long and vague every time.

Instead of…Try…
”A city at night""Slow aerial dolly forward over a city skyline at night, neon lights reflecting in wet streets, light rain"
"Ocean waves""Close-up of waves crashing on rocks in slow motion, spray catching golden sunset light"
"A person walking""A woman in a red coat walking through a rainy street, camera tracking left, shallow depth of field”

Image-to-Video: Focus on motion, not the image. The model already sees your image — describe what should move.

  • Good: “Gentle wind blowing through hair, subtle head turn to the left, soft smile”
  • Less effective: “A portrait of a woman with brown hair”

Text-to-Speech: Use punctuation to control pacing. Commas for pauses, ellipses for dramatic effect. For ElevenLabs v3, use emotion tags like [excited], [whispers], [laughs].

Sound Effects: Describe with specificity. “Thunder rumbling in the distance, followed by heavy rain on a tin roof” — not “storm sounds.”

Your prompt text carries over when you switch models. Write a prompt, switch from Kling v2.5 to Kling v3 Pro, and the text is already there. Model-specific settings reset to defaults, but your prompt stays.

  1. Adjust the prompt — add more specific details about what was wrong
  2. Try the optimizer — it may restructure your prompt more effectively
  3. Switch models — different models interpret prompts differently
  4. Adjust parameters — sometimes changing duration, resolution, or aspect ratio helps more than changing the prompt