Showcases

From a Single Line of Text to a Shot: happyhorse-1.1 Storyboard Concept Film

This piece uses the text-to-video path of happyhorse-1.1 (Alibaba) to demonstrate the minimal loop of "text → image": no first frame or reference image is provided; generation is driven solely by an English prompt containing subject, action, camera movement, lighting, and environment, verifying the model's shot comprehension and motion coherence under pure text conditions. In planning, duration is set to 8 seconds, landing in the middle of the 3–15 second range, balancing narrative completeness with per-second billing cost; resolution uses the default 1080p, and aspect_ratio is set to 16:9 for a horizontal cinematic composition. The prompt is organized into four sections—"subject → action beats → camera movement → lighting and atmosphere"—to make it easy to observe how well the model follows multi-element prompts. This format suits short-drama storyboard previsualization, advertising concept openings, and early-stage shot indications in CG workflows; since a single segment is capped at 15 seconds, multi-shot narratives require segment-by-segment generation followed by stitching.

Style Storyboard · Cinematic · Concept shot

Model HappyHorse 1.1

Try it yourself

Prompt

Try this model

Cinematic concept shot, 16:9. A lone traveler in a weathered ochre coat steps off a night ferry onto a rain-soaked wooden pier; warm lantern light spills across the wet planks while cold blue mist drifts behind. The camera begins at deck height in a low angle, then slowly rises and trucks right, keeping the traveler in the left third of the frame as the pier lights recede into the distance. Fine rain streaks catch the light, the fabric shows believable weight as it moves, and reflections ripple in shallow puddles. Shallow depth of field, soft diffused practical light sources, desaturated teal-and-amber palette, film-like grain, and gentle motion blur; steady, fluid pacing, no cuts.

Copy the full prompt and adapt it for your own generation.