Cached inputs for UCPE-conditioned Cosmos student distillation, with a short 33-frame GT seed and causal 81-frame target clips. Contains 9,188 DL3DV and 7,993 RealEstate10K scenes, 54,643 target clips, and a fixed 5+5-scene validation split. The bundle contains video VAE latents, rendered GS VAE features/masks, metric camera trajectories/intrinsics, projected text contexts, the matching projection, and provenance. Raw RGB videos and model weights are not included. Read shortseed33-v1-20261007/index.json and require complete: true and all 120 shards. A repository containing some shards is not a completed dataset. The full extracted payload is 499,974,383,360 bytes (~500 GB / 465.64 GiB)…
Independent publisher
Jinhyeok Choi
wlsgur1018
Models in Library0
Datasets in Library1
Models on Hugging Face—
Followers1
Compare the published evidence
This profile connects the 1 model and dataset record shown below to their upstream cards and sources. The library counts describe SAVRN's coverage, rather than this publisher's entire output.
Compare model specifications, source dates, and hardware estimates on the individual records. Publisher identity verification is separate from a model's quality, deployment readiness, or suitability for your work.