Robotics
Transformers
Safetensors
English
rio2
feature-extraction
Mixture of Experts
diffusion-jepa
custom_code
Instructions to use hoguai/RIO-2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use hoguai/RIO-2 with Transformers:
# pip install -U transformers accelerate # Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("hoguai/RIO-2", trust_remote_code=True, device_map="auto") - Notebooks
- Google Colab
- Kaggle
Commit History
-
Update README.md
Verified signatureb058b47 -
Update README.md
Verified signature815b0d2 -
Update README.md
Verified signature4db1eec -
Upload rio2.png
Verified signature642c7c4 -
Update README.md
Verified signature394b406 -
Upload rio2diagram.png
Verified signaturec72dd64 -
Update README.md
Verified signaturef18d312 -
Update README.md
Verified signaturedf604c2 -
Update README.md
Verified signature237c980 -
Update README.md
Verified signaturee8fbe80 -
Update README.md
Verified signature0c70dc7 -
Update README.md
Verified signaturefcd0857 -
Update README.md
Verified signature2ea893a -
Update README.md
Verified signature0ff21fe -
Update README.md
Verified signature**RIO-2** RIO-2 is a two-rate WAM(World Action Model) built for robotics. RIO-2 is composed with a low-frequency visual-language S2 backbone and a high-frequency JEPA-diffusion S1 action policy. The model is designed to separate slow scene understanding from fast robot control: • S2 refreshes visual-language context at low frequency. • Bridge/compressor modules convert S2 context into compact action-conditioning tokens. • S1 runs high-frequency action generation from cached S2 tokens and robot state. • JEPA latent prediction provides an auxiliary future-action representation. • A 10-expert S1 MoE residual path expands action capacity while keeping top-1 expert activation efficient. • This repo uses Hub custom code. Pass trust_remote_code=True until RIO-2 is merged into Transformers. RIO-2 uses JEPA-diffusion S1 action policy for general and flexible robot control in high frequency. S1 is MoE policy with 10 experts. Each expert is 100M parameter size. S2 policy is built around a we…d37d9e0 -
Upload model.safetensors
Verified signature1c38033 -
Upload 7 files
Verified signature9a5262f
-
initial commit
Verified signature1ffd957