Instructions to use ModalityDance/Omni-R1 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use ModalityDance/Omni-R1 with Transformers:
# Load model directly from transformers import AutoProcessor, AutoModelForImageTextToText processor = AutoProcessor.from_pretrained("ModalityDance/Omni-R1") model = AutoModelForImageTextToText.from_pretrained("ModalityDance/Omni-R1") - Notebooks
- Google Colab
- Kaggle
| { | |
| "crop_size": { | |
| "height": 512, | |
| "width": 512 | |
| }, | |
| "do_center_crop": true, | |
| "do_convert_rgb": true, | |
| "do_normalize": true, | |
| "do_rescale": true, | |
| "do_resize": true, | |
| "image_mean": [ | |
| 1.0, | |
| 1.0, | |
| 1.0 | |
| ], | |
| "image_processor_type": "ChameleonImageProcessor", | |
| "image_std": [ | |
| 1.0, | |
| 1.0, | |
| 1.0 | |
| ], | |
| "processor_class": "ChameleonProcessor", | |
| "resample": 1, | |
| "rescale_factor": 0.0078, | |
| "size": { | |
| "shortest_edge": 512 | |
| } | |
| } | |