gpt image prompts
GPT Image image-to-video prompt
Create a clean keyframe first, then send it to a video model with restrained camera and subject motion.
Example prompt
Keyframe for image-to-video: premium sneaker on wet asphalt, neon reflection, camera low angle, product label sharp, 16:9. Motion note: slow push-in, rain shimmer only, no product morphing.
Copy into ChatGPT Image / GPT Image; adjust product names and hex colors.
Steps
Step 1
Generate the still first
Keep the still prompt focused on composition, material, and readable product details.
Step 2
Separate motion notes
Use short motion instructions for the video step instead of asking the still prompt to animate.
Step 3
Reject morphing
If the product shape changes, reduce camera movement and ask for micro-motion only.
Multi-step pipeline: Workflow index · Recipe library
Related prompts
FAQ
- What motion is safest for product videos?
- Slow push-ins, light shimmer, steam, and fabric movement usually hold shape better than orbit shots.