1. Text only: best for creating from only words. Gen-2 will generate footage matching your prompt.
2. Image only: best for perfectly matching an image. Gen-2 will animate the image into a video. No image previews are provided since the image preview would just be your input image directly.
3. Image and text description: the video will match the image directly and use the text to help create the video.
For more information, check out Gen-2 Prompting Modes.
Text prompt tips
The following generic prompt modifiers generally produce good results:
For more action and movement, try adding keywords like:
Try using camera-specific terms:
- camera angles (
close up, etc.)
- lens types (
wide angle, etc.)
- camera movements (
- In general, trying to explain camera movements as if you were explaining to a toddler can help; so instead of
dolly shot, you would say something like
zooming in over timeor
growing over time
- This structure works well: [base prompt] in the style of [style], [aesthetic]
first person viewto improve POV results
- Try experimenting with emojis! 🐠🌸🌟
- Don’t be afraid to ask for something abstract and see what you get