How to Write Prompts for an AI Video Generator
A practical guide to writing text-to-video and photo-to-video prompts in VidStride: subject, motion, camera, lighting and format.
What VidStride generates
VidStride turns a written prompt, or a photo plus a prompt, into one 5-second video at 720p. Text to video is available in landscape 16:9 and vertical 9:16. Photo to Video follows the framing of the photo you upload.
Each generation uses the credits shown on the tool before you start. If the video service clearly reports a failure, those credits are returned.
Start with one clear subject
Name the main subject first and keep it to one or two characters or objects. "A golden retriever" works better than a long list of things that should all appear at once.
Add the details that matter for recognition: color, clothing, age, material or style.
Describe the motion, not just the scene
Five seconds is short, so describe one main action: running through surf, turning toward the camera, leaves falling. Several unrelated actions in one prompt are often blended or dropped.
Use speed words such as slow motion, gentle or fast to set the pace.
Direct the camera and light
Camera terms help: close-up, wide shot, low tracking shot, aerial drone shot, slow push-in. Pick one camera move per clip.
Lighting and mood words such as golden hour, neon night, soft studio light or misty morning strongly shape the result.
Tips for Photo to Video
Use a clear, well-lit photo where the subject is fully visible. Then describe how it should move, for example "smiles and waves at the camera" or "hair moves in the wind".
Only upload photos you have the right to use. Do not upload images of other people without their permission.
Choose the right format
Use 16:9 for YouTube, websites and presentations. Use 9:16 for TikTok, Instagram Reels and YouTube Shorts.
Finished videos are saved to your library so you can download them again later.