Amazon's Seller Assistant drafts ads: animate a product photo
Amazon's Seller Assistant checks listings and drafts Sponsored campaigns. Prepare the creative: animate your real product photo with Sume's frame_images.

If Amazon's Seller Assistant is going to draft your advertising campaigns, give it good creative to point at. With Sume you can animate a real product photo into a short clip by sending it as the first frame in frame_images on POST /v1/videos, and keep the item itself unchanged.
Amazon's Accelerate 2026 recap says Seller Assistant for Advertising evaluates whether products are ready to advertise by checking listings and inventory, then drafts Sponsored Products and Sponsored Brands campaigns for the seller's approval (read 2026-10-04). The same page does not mention image or video generation tools.
What does Amazon's page say about ads?
Only what is on the recap. Read the vendor page for terms and availability.
| Item | What the page says |
|---|---|
| Seller Assistant for Advertising | Checks listings and inventory for ad readiness; drafts campaigns for approval |
| Auto-optimization | Adds performing search terms, pauses underperformers within seller bid limits; new campaigns saw a first sale 40 percent faster |
| Ads Data Manager | Free tool connecting first-party customer data; 30 percent better returns on ad spend |
| Creative generation | Not described on the page |
How does the first-frame approach work?
Per the video docs, frame_images holds first or last frame images for image-to-video; each entry carries a frame_type of first_frame or last_frame. If both frame_images and input_references are sent, frame_images takes precedence and the request is treated as image-to-video. Ask only for camera or light movement, not a different product.
import os, requests
r = requests.post(
"https://api.sume.com/v1/videos",
headers={"Authorization": "Bearer " + os.environ["SUME_API_KEY"]},
json={
"model": "seedance-2.5",
"prompt": "Slow camera push-in, soft light moves across the product, nothing else changes",
"frame_images": [
{
"type": "image_url",
"image_url": {"url": "https://example.com/product.png"},
"frame_type": "first_frame",
}
],
"duration": 6,
"resolution": "720p",
"aspect_ratio": "9:16",
},
timeout=60,
)
print(r.status_code, r.json().get("polling_url"))
What do you check before it goes into a campaign?
Play it frame by frame. Look for the product changing shape, labels distorting or parts appearing. If the product drifts, shorten the clip or simplify the motion. Advertising rules for video and AI-generated content are set by Amazon; read them on Amazon's own pages before you upload.
Keep the job id with the SKU so the creative behind each campaign can be found again.
Why not let the assistant decide?
Amazon says it drafts campaigns for your approval, and that is where your review of the creative belongs. The model sees the photo and a prompt; you know which claims are true.
Sources
Related posts
More in Use cases
- Amazon multichannel tools: one media set for four channels
Seller Central now links eBay, Shopify, TikTok Shop and Walmart. Auto-formatting is listed as upcoming; media is still yours. Plan one set with Sume.
- Amazon Sponsored Brands video: 1 Mbps and 96 kbps audio minimums
Amazon Sponsored Brands video needs at least 1 Mbps video, 96 kbps audio at 44.1 kHz, and 16:9 only. What to check on an AI-generated clip before upload.
- Anniversary video from old photos: three 10-second clips and a song
Animate three old photos with Gemini Omni image_to_video, 10 seconds each, then join them over a Lyria 3.5 track in Timeline. 30 seconds bills at $0.10.
- App preview clip from screenshots with Seedance 2.5 and a caption pass
Screenshots as reference images can drive a 15 s Seedance 2.5 app promo on Sume for $8.67 at 720p. Generated UI text is unreliable, so add words after.
Written by Sume