motus-robotics/Motus_Wan2_2_5B_pretrain

Robotics Model

This is a text-to-video generation model from Motus Robotics, based on the WAN architecture, designed for robotics manipulation tasks. It is a Stage-1 pretrained model that generates videos from language instructions, using the transformers library and safetensors format.

Catalog metadata

Repository owner

motus-robotics

Source

Hugging Face Models

Task

text to video

Cameras

0

License

apache-2.0

Browse the catalog to compare robots, tasks, sensors, and licenses. Sign in to open protected download and model links.