Instructions to use Nvidia-CMU25/DiffusionVideo2WorldGeneration with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use Nvidia-CMU25/DiffusionVideo2WorldGeneration with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("feature-extraction", model="Nvidia-CMU25/DiffusionVideo2WorldGeneration", trust_remote_code=True)# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("Nvidia-CMU25/DiffusionVideo2WorldGeneration", trust_remote_code=True, dtype="auto") - Notebooks
- Google Colab
- Kaggle
Fixed pipeline call
Browse files- video2world_hf.py +1 -1
video2world_hf.py
CHANGED
|
@@ -104,7 +104,7 @@ class DiffusionVideo2World(PreTrainedModel):
|
|
| 104 |
continue
|
| 105 |
|
| 106 |
# Generate video
|
| 107 |
-
generated_output = pipeline.generate(
|
| 108 |
prompt=current_prompt,
|
| 109 |
image_or_video_path=current_image_or_video_path,
|
| 110 |
negative_prompt=cfg.negative_prompt,
|
|
|
|
| 104 |
continue
|
| 105 |
|
| 106 |
# Generate video
|
| 107 |
+
generated_output = self.pipeline.generate(
|
| 108 |
prompt=current_prompt,
|
| 109 |
image_or_video_path=current_image_or_video_path,
|
| 110 |
negative_prompt=cfg.negative_prompt,
|