ARDY Motius Artifact

Load with ARDYPipeline.from_pretrained(...). The ARDY checkpoint is stored in this artifact. The official text encoder uses gated LLM2Vec/Llama repositories; make sure the runtime has a Hugging Face token with access to meta-llama/Meta-Llama-3-8B-Instruct.

from motius.pipelines.ardy import ARDYPipeline

pipe = ARDYPipeline.from_pretrained("<artifact-path-or-repo>", bundle_kwargs={"device": "cuda"})
out = pipe.text_to_motion("a person walks forward and waves.", 120, num_denoising_steps=4)

Direct Loading

from motius import Pipeline

pipeline = Pipeline.from_pretrained("ZeyuLing/motius-ardy-330-horizon40")
Downloads last month

-

Downloads are not tracked for this model. How to track
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support