Pretrained DiT decoders for
TT-VidT
(
Decoupling the Temporal Axis for Efficient Motion-Centric Video Pretraining
,
NeurIPS 2026). Each TT-VidT encoder pretraining run initialises its decoder from one of
these; the training code downloads them automatically through
DECODER_PRETRAINED
,
e.g.
"KBlueLeaf/TTVidT-decoders/pretrain_video_S_qknorm"
.
The decoders work in the 32x32x4 latent space of the frozen f8 frame VAE
KBlueLeaf/latentmaid-vae
and are
conditioned on DINOv3 ViT-B/16 features (768-d).
ImageNet
decoders reconstruct an ImageNet image from its DINOv3 class token.
Video
decoders reconstruct a later video frame from its class token plus, through
cross-attention, the spatial features of an earlier frame.
Files
Each decoder is
<name>.safetensors
plus
<name>.json
, which holds the decoder
config (
decoder_config
), the latent size and the latent statistics used in training.
TTVidT-decoders huggingface.co is an AI model on huggingface.co that provides TTVidT-decoders's model effect (), which can be used instantly with this KBlueLeaf TTVidT-decoders model. huggingface.co supports a free trial of the TTVidT-decoders model, and also provides paid use of the TTVidT-decoders. Support call TTVidT-decoders model through api, including Node.js, Python, http.
TTVidT-decoders huggingface.co is an online trial and call api platform, which integrates TTVidT-decoders's modeling effects, including api services, and provides a free online trial of TTVidT-decoders, you can try TTVidT-decoders online for free by clicking the link below.
KBlueLeaf TTVidT-decoders online free url in huggingface.co:
TTVidT-decoders is an open source model from GitHub that offers a free installation service, and any user can find TTVidT-decoders on GitHub to install. At the same time, huggingface.co provides the effect of TTVidT-decoders install, users can directly use TTVidT-decoders installed effect in huggingface.co for debugging and trial. It also supports api for free installation.