HappyHorse 1.1 Image-to-Video: Core Parameters and Use Cases Illustrated
A real sample short clip generated with the HappyHorse 1.1 (Alibaba) image-to-video pipeline, demonstrating the model's camera movement and subject performance when driven by a single first-frame image. The schema shows that this pipeline requires only one first-frame image (image_urls, exactly 1 image), an optional prompt to describe motion and style, duration defaulting to 5 seconds with a range of 3–15 seconds, and resolution defaulting to 1080p with 720p optional. This piece uses a product first-frame image of lemon soda as the sole required asset, with 5-second, 1080p output; the prompt focuses on a slow push-in and the motion of liquid bubbles, illustrating the division of labor where "the first-frame image determines composition and subject, and the prompt determines motion." Applicable scenarios include e-commerce advertising and product showcase short videos, brand marketing assets, short social media clips, and preview clips in CG and creative workflows. It is suitable for creating short-duration image-to-video content with this model family; a single segment is capped at 15 seconds and 1080p maximum, and long narratives require stitching multiple segments.
Style Image-to-Video · Product Image · Character Consistency
Model HappyHorse 1.1
Reference inputs

image_urls
Prompt
Slowly and steadily push in toward a glass of iced sparkling lemon soda placed on a wet marble countertop, with condensation droplets continuously forming and sliding down the glass wall, fine bubbles persistently rising in the liquid, and a thin lemon slice gently turning at the rim of the glass. Shallow depth of field, soft warm backlight glowing through the drink, cool-toned studio fill light on the rim of the glass, bringing out a subtle chilled mist. The camera moves forward at a constant slow speed with no shake, from macro to medium shot composition, high temporal consistency between frames, realistic commercial product photography, 16:9, 1080p.
Copy the full prompt and adapt it for your own generation.
Tags