Inputs:
- Text prompt
- Start image
- End image (optional)
- Image references (optional)
- Elements
Parameters
- Duration
- Native audio
- Prompt Strength
Limits:
- Duration: 3–15 seconds
- Output: up to 1080p
- Aspect ratio depends on the selected endpoint/settings
- Non-English prompts or audio may be translated into English automatically
Tips:
- Use Standard for quick iterations, previews, b-roll, and cheaper shot exploration
- Use reference elements for better character and object consistency
- Split complex scenes with multiple characters or difficult choreography into separate shots
- Define the action clearly in the prompt: camera movement, pacing, lighting, and mood