Nano Banana 2.1: prepare a first frame for AI video
Written by Kadenly · Updated
Prepare a readable character or product image externally, then animate the exported image in Image to Video.
Purpose and limitations
This image model is not offered as a generator on this site. Prepare and export the still externally, then use Image to Video here. Official capability descriptions are not our test results.
Edit the still before adding motion
Google positions Nano Banana 2.1 as an image generation and editing model. Use it to settle the subject, framing and background before trying to animate them. It does not produce a video in this workflow.
Export the frame you actually want
Leave room around the subject and inspect the face, fingers and lettering at full size. Multi-image references may help preparation, but a clean final still is what you upload to the video tool.
Write motion instead of redescribing the image
Upload the exported image to Image to Video and start with one small action and one camera move. Check the selected model options and quote. Still-image consistency does not guarantee consistency throughout a video.
English prompt example
This is an editable English example, not a verified generation result. Adapt it to your actual photo and casting.
A subtle breeze moves the clothing while the camera slowly pushes in. Keep the same person, outfit, background and framing. No new objects, no dialogue. One continuous shot.Input and result checklist
- Inspect the subject, fingers, lettering and frame edges.
- Match every @Image mention to the actual material.
- Check model, duration, aspect ratio and quote before submitting.
- Play the video and review its first and last frames, not just the thumbnail.
Common issues and adjustments
Identity or product details change
Use a clearer reference, reduce subject and camera movement, and keep the subject visible. References and prompts do not guarantee unchanged details.
Sources and related policies
These are original preparation tips and prompt examples from our team. Model capabilities change; check current settings before generating. For input requirements and usage rules, see: