GivingTuesday 2026 video ideas: a captioned 3-scene ask

GivingTuesday 2026 falls on December 1, 2026, per givingtuesday.org (read 2026-09-30). A short ask works as three scenes: a hook, one impact fact, and a call to action. Sume's avatar video takes ordered video_inputs for exactly that, accepts a total of 4 to 60 seconds, and can burn captions into the final MP4.

The date and the site's "Generosity Toolbox" are from the GivingTuesday homepage, which has no AI-specific announcement. Sume details are from Avatar videos, Music 1.0 and Timeline 1.0. Write your own facts and numbers in the script.

How do I structure the three scenes?

Use video_inputs instead of a single script. Each spoken scene is type: "text" with one of script or input_text; a silence beat uses voice.type: "silence" with a required duration. Launch requests reference a ready avatar with avatar_handle, and total planned duration must land in 4 to 60 seconds. Execution supports one avatar per final video.

How do captions and the music bed fit in?

Optional captions on the avatar request burns styles into the clean final MP4 after generation, from the spoken text. Inline captions are rejected when the estimated duration is above 60 seconds, and a caption failure soft-fails: the video can still succeed with captions.status=failed. For a music bed, make a track with Music 1.0 and place it as the Timeline soundtrack; duck_db of 0 to 20 lowers it under the voice and needs a real audio spine, not silence. See Timeline duck music under voiceover.

What does one finished cut cost?

Rates come from the pricing page rows. Avatar video is billed per second, so multiply by your script's length.

Which format should I post?

Avatar video supports 1:1, 3:4, 9:16, 4:3 and 16:9, defaulting to 9:16 at 720p; see avatar video aspect ratios. For more nonprofit ideas, read fundraising video ideas for nonprofits.