MoTok
Official weights for MoTok: Learning Structured Tokenization for Human Motion Representation.
Contents
t2m-ft-from-t5-base/: Text-to-Motion finetuned T5 checkpoint (inference files only).
Load T2M model
from transformers import T5Tokenizer, T5ForConditionalGeneration
model_id = "AIGeeksGroup/MoTok"
sub = "t2m-ft-from-t5-base"
tokenizer = T5Tokenizer.from_pretrained(model_id, subfolder=sub)
model = T5ForConditionalGeneration.from_pretrained(model_id, subfolder=sub)
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support
# pip install -U transformers accelerate # Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("AIGeeksGroup/MoTok", device_map="auto")