Our Gen-2 model has three unique ways of prompting: text only, image only, and image plus text. Curious about how they differ and what results each prompting mode will create? We've outlined the basics below:
Text Only (Text to Video)
In this mode, you only have to enter a text prompt describing what you'd like a clip of. For help with text prompting, check out our Gen-2 Prompting Guide.
💡Previews are enabled.
Image Only (Image to Video)
In this mode, the generated output video will most closely resemble the image input, as if the input image is the "first frame" of the generated video.
💡 Previews are currently disabled in this mode, and the "Free Preview" button will remain greyed out. The image you input is the only preview you'll need.
Image Plus Text (Image to Video With Text Reference)
In this mode, the video will match the image directly and use the text to help create the video.
💡 Previews are also currently disabled in this mode.
Frequently Asked Questions
Q: What aspect ratios are supported in each of these three modes?
A: Text Only only currently supports horizontal aspect ratios. You can use an input of any aspect ratio you'd like, but you will always receive a horizontal result. Image Only and Image Plus Text currently allows for input and export of any export aspect ratio you'd like — for instance, if you were to use a vertical input, you'd receive a vertical output of the same aspect ratio.