Any-to-Any
Transformers
Safetensors
PyTorch
NemotronH_Nano_Omni_Reasoning_V3
feature-extraction
nvidia
multimodal
custom_code
Instructions to use Jashan887/76_Nvidia_Reasoning_30B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use Jashan887/76_Nvidia_Reasoning_30B with Transformers:
# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("Jashan887/76_Nvidia_Reasoning_30B", trust_remote_code=True, dtype="auto") - Notebooks
- Google Colab
- Kaggle
| { | |
| "image_processor_type": "NemotronH_Nano_Omni_Reasoning_V3ImageProcessor", | |
| "auto_map": { | |
| "AutoImageProcessor": "image_processing.NemotronH_Nano_Omni_Reasoning_V3ImageProcessor", | |
| "AutoVideoProcessor": "video_processing.NemotronH_Nano_Omni_Reasoning_V3VideoProcessor", | |
| "AutoProcessor": "processing.NemotronH_Nano_Omni_Reasoning_V3Processor" | |
| }, | |
| "patch_size": 16, | |
| "downsample_ratio": 0.5, | |
| "norm_mean": [0.48145466, 0.4578275, 0.40821073], | |
| "norm_std": [0.26862954, 0.26130258, 0.27577711], | |
| "min_num_patches": 1024, | |
| "max_num_patches": 13312, | |
| "max_model_len": 16384 | |
| } | |