by FAL AI (Kling 4K)
Kling O3 4K — reference-to-video with @Element character locking at native 4K. Up to 7 refs (3-15s)
Detailed content for this model is coming soon.
FAL AI (Seedance)
ByteDance Seedance 2.0 via FAL — multimodal ref system: up to 9 images + 3 videos + 3 audio, native audio
Kling Direct
Kling V3 Standard via direct API — 720p text-to-video (5/10/15s)
Kunya (Seedance)
ByteDance Seedance 2.0 Mini — low-cost multimodal @-reference: up to 9 images + 3 videos + 3 audio