Christmas Jumper Day 2026 (Dec 10): a fundraiser clip from one photo

Christmas Jumper Day 2026 is Thursday 10 December, so a workplace or school that wants a reminder clip has about eight weeks. The cheapest build on Sume is one team or classroom photo turned into a short image-to-video clip, then a second job that burns the date and donation line onto it as caption cues. The clip is silent, so the cues are authored text, not speech-to-text.

The date and the suggested amounts come from Save the Children UK's own page, read on 2026-10-12. Everything about Sume below comes from the docs pages listed under sources.

What does Save the Children say about Christmas Jumper Day 2026?

The page gives Thursday 10 December 2026 as the date and says people typically donate 2 pounds for adults and 1 pound for children. It also says groups can pick a different day if another one suits a school or workplace. For workplaces it describes a simple route, an email with a JustGiving link and a reminder to wear jumpers, and mentions bucket collections and employer matching. For schools it lists posters, assembly presentations and lesson plans, with collection either in person or through JustGiving.

Your clip should repeat only what you know: your own date, your own collection method and your own link. If your group picks another day, put that day in the cues, not 10 December.

Which Sume calls build the clip?

Step one is a video job at POST /v1/videos with frame_images holding your photo as a first_frame. Sume treats that as image-to-video. Before you pick a model, call GET /v1/videos/models and read supported_frame_images, supported_durations and supported_aspect_ratios for it, because limits differ by model. The docs state, for example, that wan-3.0 accepts 2 to 30 seconds.

Step two is POST /v1/video-captions with the finished clip's public HTTPS URL as video_url and a cues array of text, start and end in seconds. Because you send cues, Sume does not run speech-to-text and burns your text at those times. That is the documented path for silent clips. A clip with no audible speech and no cues fails with caption_no_speech, so do not skip the cues.

What should the cues say?

Keep each cue short enough to read in two seconds. A five-cue plan for a 10-second clip:

How do you send the caption job?

The request below builds the cue payload in Python and prints it, so you can check the JSON before you post it with your own API key and clip URL.

import json

cues = [
    (0, 2, "Jumper Day is Thursday 10 December"),
    (2, 4, "Wear your best (or worst) jumper"),
    (4, 6, "Suggested: 2 pounds adults, 1 pound children"),
    (6, 8, "Give at the link in this post"),
    (8, 10, "See you Thursday"),
]
payload = {
    "video_url": "https://example.com/jumper-day-clip.mp4",
    "cues": [{"text": t, "start": s, "end": e} for s, e, t in cues],
}
print(json.dumps(payload, indent=2))

What can go wrong?

The video URL must be public HTTPS, so host the finished clip before the caption job. Watch the jumper itself in the photo: generated motion can change patterns and any lettering, so look at the result before you caption it. Last, the amounts are the charity's suggestion for its own campaign. If you run a local collection, state your own figure instead.