Robotics
Transformers
PyTorch
English
RDT
rdt
RDT 2
Vision-Language-Action
Bimanual
Manipulation
Zero-shot
UMI
Flowmatching
Diffusion
Action Expert
Instructions to use robotics-diffusion-transformer/RDT2-FM with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use robotics-diffusion-transformer/RDT2-FM with Transformers:
# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("robotics-diffusion-transformer/RDT2-FM", dtype="auto") - Notebooks
- Google Colab
- Kaggle
| { | |
| "act_adaptor": "mlp3x_silu", | |
| "act_pos_emb_config": [ | |
| [ | |
| "action", | |
| 24 | |
| ], | |
| [ | |
| "register", | |
| 4 | |
| ] | |
| ], | |
| "action_dim": 20, | |
| "img_pos_emb_config": null, | |
| "img_token_dim": 2176, | |
| "lang_pos_emb_config": null, | |
| "lang_token_dim": null, | |
| "max_img_len": null, | |
| "max_lang_len": null, | |
| "noise_scheduler": { | |
| "num_inference_timesteps": 5 | |
| }, | |
| "pred_horizon": 24, | |
| "rdt": { | |
| "depth": 14, | |
| "ffn_dim_multiplier": null, | |
| "hidden_size": 1024, | |
| "multiple_of": 256, | |
| "norm_eps": 1e-05, | |
| "num_heads": 8, | |
| "num_kv_heads": 4, | |
| "num_register_tokens": 4, | |
| "use_flash_attn": true | |
| }, | |
| "selected_layers": [ | |
| 0, | |
| 1, | |
| 2, | |
| 3, | |
| 4, | |
| 5, | |
| 6, | |
| 7, | |
| 8, | |
| 9, | |
| 10, | |
| 11, | |
| 12, | |
| 13 | |
| ], | |
| "state_adaptor": "mlp3x_silu", | |
| "state_dim": 20 | |
| } |