Kandinsky 6.0 lite-distill
Kandinsky 6.0 lite-distill from Kandinsky6: text-to-video and image-to-video; Planned.
Kandinsky 6.0 lite-distill is a video generation model from Kandinsky6. The catalog records it for text-to-video, image-to-video, and audio-video-generation.
No official teaser or method figure is checked into this docs tree.
Code: kandinskylab/kandinsky-6.
Build your run
Kandinsky6Pipelineworldfoundry.pipelines.kandinsky6.pipeline_kandinsky6worldfoundry.pipelineworldfoundry-eval zoo model-download --model-id kandinsky6-lite-distill --check-local --jsonWhat it is for
Use kandinsky6-lite-distill for the family's joint T2V/I2V inference path. The catalog records both task profiles on this page.
Catalog tasks include text-to-video, image-to-video, and audio-video-generation. Documented modalities: text and image in, video out.
Runtime contract
Optional prompt, image, seed, frames, and fps; artifacts generated_video and kandinsky6-lite-distill.mp4.
CLI task profiles: text-to-video, image-to-video, audio-video-generation, and video_generation. WorldFoundry binds Kandinsky6Pipeline.
Family status: Planned.
Native audio-video inference; checkpoint-backed CUDA parity is pending.
A pipeline binding is recorded, but a successful run has not been established for this entry.
Running in WorldFoundry
The catalog launch is worldfoundry-eval run kandinsky6-lite-distill --pipeline.task-profile text-to-video. The command builder above switches --pipeline.task-profile among recorded tasks and still emits worldfoundry-eval run.
It runs in the unified environment worldfoundry-unified-cu128 (Python 3.11, CUDA 12.8).
Recorded weight repositories include kandinskylab/Kandinsky-6.0-Lite-distill-5s-Diffusers. Prepare the environment and check local assets with bash scripts/inference/prepare_model_infer.sh kandinsky6-lite-distill. Add --download to fetch missing public checkpoints.
Status: Planned.
