Snapchat AI Clips make 5-second videos; do it outside Snapchat
Snap's AI Clips turn a photo into a five-second video inside a Lens. Here is how to get the same photo-to-clip result as a file you can post anywhere.

Snapchat's AI Clips are Lenses that turn a Snap or Camera Roll photo into a five-second AI video, and they run inside Snapchat for Lens+ subscribers. If you want the same photo-to-five-second-clip result as a file you can post on any channel, send the photo to Sume's video API as a first frame: Seedance 2.5 accepts 4 to 30 seconds, so five seconds is inside its range, and the job returns an MP4 you own.
Snap's facts here come from its newsroom announcement, read on 2026-10-03; Sume's from its video generation docs. The two are different products: AI Clips is a distribution-ready Lens, while Sume is a generation API that does not publish anywhere.
What each side offers
The comparison is about where the result lives and who writes the prompt, not about which looks better; judge quality on your own photos.
| Question | Snapchat AI Clips | Sume /v1/videos |
|---|---|---|
| Input | A Snap or Camera Roll photo | A public HTTPS image as a first frame |
| Length | Five seconds | Seedance 2.5: 4 to 30 seconds |
| Who writes the prompt | The Lens creator, a closed prompt | You, per request |
| Where it runs | Inside Snapchat, for Lens+ subscribers | An API; you download the MP4 |
| Built with | Lens Studio's GenAI Suite | One HTTP call and a poll |
Send one photo as a first frame
On /v1/videos, an image in frame_images with frame_type set to first_frame means image-to-video. Ask for a small motion, such as hair moving, steam rising or a slow push in, and request five seconds. Poll the job until it completes, then download the content URL.
curl -X POST https://api.sume.com/v1/videos \
-H "Authorization: Bearer $SUME_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "seedance-2.5",
"prompt": "Slow push in. Steam rises from the cup. Nothing else moves.",
"duration": 5,
"resolution": "720p",
"aspect_ratio": "9:16",
"frame_images": [{
"type": "image_url",
"image_url": {"url": "https://example.com/photo.jpg"},
"frame_type": "first_frame"
}]
}'Make a batch and a template
A Lens fixes the prompt so a user only supplies a photo. You can copy that discipline: write one prompt per effect, keep a list of photos, and loop over them from your own code. Several five-second clips then join in Timeline with a music bed for a post that runs 15 to 30 seconds.
Honest limits
Sume has no one-tap experience for your followers, no Lens, and no Snapchat distribution. Photos of other people need their permission, and the photo URL must be public HTTPS, not a private link. Check each clip at full size, since small details such as hands and text can drift.
Sources
Related posts
More in Use cases
- Snap DPA for Sponsored Snaps: catalog images to fix with Sume
Snap's Dynamic Product Ads now run in Chat for all global advertisers. The creative comes from your catalog, so the work is catalog images Sume can fix.
- Snapchat Dynamic Story ads: 360x600 preview tile, 993x284 logo
Snap Dynamic Story ads show 3 to 10 catalog products from a Discover tile: 360x600 PNG, 993x284 logo, 55-character headline. Make the tile with Sume.
- Snapchat product-level video ads: one video per product, 1080x1920
Snap product-level video for Dynamic Product Ads: one video per product, 1080x1920 minimum, 3 to 180 s, 1 GB, with an image fallback. Make the clips with Sume.
- Spain's AI bill and AI labels: in Congress, not law yet
Spain's AI governance bill reached Congress in June 2026 with fines up to 35 million euros. Is AI video labelling Spanish law yet, and what to check on a file?
Written by Sume