Documentation Index
Fetch the complete documentation index at: https://docs.loraverse.io/llms.txt
Use this file to discover all available pages before exploring further.
Each video model accepts different inputs and rewards different prompting. The Bench Bar adapts to whichever you pick.
First-frame
First-and-last-frame
Omni-reference
Multi-shot
You give the model one image, and it generates motion outward from there. The hero frame of your shot becomes the starting frame. The prompt directs the motion.
Two anchor images bracketing the take. The model fills in the motion between them. Useful when you know where the shot starts and ends, and you want continuity.
Models like Seedance 2.0 and Kling can take a richer mix. Cast images, environment refs, multiple panels. The Bench Bar surfaces more attachment slots when you pick these.
Multiple cuts within a single render, with per-shot timings. When supported, the Bench Bar shifts into shot-mode. Each shot gets its own description and duration. The model handles transitions.
