README.md

---
license: mit
library_name: transformers
tags:
- robotics
- vla
- diffusion
- multimodal
- pretraining
language:
- en
pipeline_tag: robotics
---
# CogACT-Large

CogACT is a new advanced VLA architecture derived from VLM. Unlike previous works that directly repurpose VLM for action prediction by simple action quantization, we propose a componentized VLA architecture that has a specialized action module conditioned on VLM output. CogACT-Large employs a [DiT-L](https://github.com/facebookresearch/DiT) model as the action module.

All our [code](https://github.com/microsoft/CogACT), [pretrained model weights](https://huggingface.co/CogACT), are licensed under the MIT license.

Please refer to our [project page](https://cogact.github.io/) and [paper](https://arxiv.org/abs/2411.19650) for more details.


## Model Summary

- **Developed by:** The CogACT consisting of researchers from [Microsoft Research Asia](https://www.microsoft.com/en-us/research/lab/microsoft-research-asia/).
- **Model type:** Vision-Language-Action (language, image => robot actions)
- **Language(s) (NLP):** en
- **License:** MIT
- **Model components:**
  + **Vision Backbone**: DINOv2 ViT-L/14 and SigLIP ViT-So400M/14
  + **Language Model**: Llama-2
  + **Action Model**: DiT-Large
- **Pretraining Dataset:** A subset of [Open X-Embodiment](https://robotics-transformer-x.github.io/)
- **Repository:** [https://github.com/microsoft/CogACT](https://github.com/microsoft/CogACT)
- **Paper:** [CogACT: A Foundational Vision-Language-Action Model for Synergizing Cognition and Action in Robotic Manipulation](https://arxiv.org/abs/2411.19650)
- **Project Page:** [https://cogact.github.io/](https://cogact.github.io/)

## Uses
CogACT takes a language instruction and a single view RGB image as input and predicts the next 16 normalized robot actions (consisting of the 7-DoF end effector deltas
of the form ``x, y, z, roll, pitch, yaw, gripper``). These actions should be unnormalized and integrated by our ``Adaptive Action Ensemble``(Optional). Unnormalization and ensemble depend on the dataset statistics.

CogACT models can be used zero-shot to control robots for setups seen in the [Open-X](https://robotics-transformer-x.github.io/) pretraining mixture. They can also be fine-tuned for new tasks and robot setups with an extremely small amount of demonstrations. See [our repository](https://github.com/microsoft/CogACT) for more information.

Here is a simple example for inference.

```python
# Please clone and install dependencies in our repo
# Install minimal dependencies (`torch`, `transformers`, `timm`, `tokenizers`, ...)

from PIL import Image
from vla import load_vla
import torch

model = load_vla(
      'CogACT/CogACT-Large',
      load_for_training=False,
      action_model_type='DiT-L',
      future_action_window_size=15,
    )                                 
# about 30G Memory in fp32; 

# (Optional) use "model.vlm = model.vlm.to(torch.bfloat16)" to load vlm in bf16

model.to('cuda:0').eval()

image: Image.Image = <input_your_image>
prompt = "move sponge near apple"           # input your prompt

# Predict Action (7-DoF; un-normalize for RT-1 google robot data, i.e. fractal20220817_data)
actions, _ = model.predict_action(
          image,
          prompt,
          unnorm_key='fractal20220817_data', # input your unnorm_key of dataset
          cfg_scale = 1.5,                   # cfg from 1.5 to 7 also performs well
          use_ddim = True,                   # use DDIM sampling
          num_ddim_steps = 10,               # number of steps for DDIM sampling
        )

# results in 7-DoF actions of 16 steps with shape [16, 7]
```

## Citation

```bibtex
@article{li2024cogact,
  title={CogACT: A Foundational Vision-Language-Action Model for Synergizing Cognition and Action in Robotic Manipulation},
  author={Li, Qixiu and Liang, Yaobo and Wang, Zeyu and Luo, Lin and Chen, Xi and Liao, Mozheng and Wei, Fangyun and Deng, Yu and Xu, Sicheng and Zhang, Yizhong and others},
  journal={arXiv preprint arXiv:2411.19650},
  year={2024}
}
}
```