NOVA Video

by Beijing Academy of Artificial Intelligence

About the model:

Autoregressive video generation model.

Model type:

Video

Model performance class:

Limited

Link to the model:

https://huggingface.co/BAAI/nova-d48w1024-osp480

Base models:

(undefined)

End model:

NOVA-D48W1024-OSP480

End model license:

Apache-2.0

About the organisation:

Chinese non-profit AI lab.

Link to the organisation:

https://www.baai.ac.cn/en/

Model release date:

December 2024

Availability

Base Model Data

web datasets LAION-5B and COYO-700M

https://huggingface.co/BAAI/nova-d48w1024-osp480

End User Model Data

web datasets LAION-5B and COYO-700M

https://huggingface.co/BAAI/nova-d48w1024-osp480

Base Model Weights

https://huggingface.co/BAAI/nova-d48w1024-osp480/tree/main

End User Model Weights

https://huggingface.co/BAAI/nova-d48w1024-osp480/tree/main

Training Code

https://huggingface.co/BAAI/nova-d48w1024-osp480/tree/main

Documentation

Code Documentation

https://huggingface.co/BAAI/nova-d48w1024-osp480/tree/main

Hardware Architecture

https://huggingface.co/BAAI/nova-d48w1024-osp480

Preprint

https://arxiv.org/abs/2412.14169

Paper

Model published through ICLR 2025 poster.

https://openreview.net/forum?id=JE9tCwe3lp

Modelcard

https://huggingface.co/BAAI/nova-d48w1024-osp480

Datasheet

Access

Package

API and Meta Prompts

Licenses

weights and code under apache 2

Is this information not up to date?

Contribute here ->

Supported by the Centre for Language Studies and the Dutch Research Council. Website design & development © 2024 by BSTN. This version of the index generated 10 March 2026, website content last updated 11 March 2026.