AI Hug Video Generator
Turn a shared photo into a gentle hug video
Animate one photo already showing two consenting adults into a short AI hug video. Describe a gentle embrace, review the plan, then approve generation.
Open the video Agent
Create your video
Source images
- Shared photo of two adults
Upload one photo
Estimated generation: 20 credits
Credits are used only when generation starts.
Start with one shared photo
Both adults must already be in the same photo, with faces and upper bodies visible. This tool does not merge two separate portraits. A little space around their arms gives the movement room.
Try a small, natural gesture
A brief embrace can make a shared memory feel warmer. Avoid asking for several actions at once. Hands, faces and clothing may shift; inspect the clip before sharing it.
How to make your video
1. Choose clear photos
Upload the required photos with a clear subject and an uncluttered background. Use photos you have permission to animate.
2. Describe the movement
Tell the Agent what should move and what should stay still. Keep the action simple; complex movement can distort details.
3. Review and confirm
Read the plan and its credit quote before confirming. Every new generation, including a revised version, needs a new approval.
Example: photos to motion
Illustrative Grok example made from these source photos, not a production-model quality guarantee. Play to see the movement; the still frame alone cannot show it.


Questions before generating
Can I upload two separate portraits?
No. Upload one photo that already contains both adults. Combining separate portraits is not part of this tool.
Is generation free?
No. Video generation uses credits. The current server quote is 20 credits per video; you must approve it before generation starts.
Can I revise an existing video?
Describe a new version to the Agent. It generates another video from your original photos and revised instructions, not by editing the MP4. A new quote and approval are required.
Can I control speech or audio?
This tool does not promise speech, lip sync or a particular soundtrack. Describe visual motion rather than relying on audio.