A 30-second video generated from a single image and one prompt (prompt included)

umesh_ai · x · 2026-09-25

User umeshai shared a 30-second video generated entirely from a single image and one prompt, with the full prompt attached. The demo showcases current image-to-video models' ability to sustain multi-scene continuous narration, and the prompt is available for anyone to reproduce and test.

Related event: Single Image Plus One Prompt Generates 30-Second Cinematic Video(2 posts)→

Original post →

More from Multimodal

Multimodal channel →