Seedance 2.5 lands on fal
By Steven Van ·
The model turns a still image with a hand-drawn path into a continuous, keyframe-free video, per fal's own demo.
Seedance 2.5 is now available on fal. In fal's demo, a single still image with a red squiggle drawn on it becomes a continuous first-person run through a torii gate and a full lap around a monorail, generated without any manual keyframing.
fal
A generative media platform and inference API for running fast image, video, and audio AI models at scale.
View fal →