Any-to-Any
Transformers
Safetensors
NemotronH_Nano_Omni_Reasoning_V3
feature-extraction
multimodal
Merge
nvfp4
mixture-of-experts
speculative-decoding
custom_code
8-bit precision
modelopt
Instructions to use brazley/Brazley-Nemotron-3.5-Lightning-Omni with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use brazley/Brazley-Nemotron-3.5-Lightning-Omni with Transformers:
# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("brazley/Brazley-Nemotron-3.5-Lightning-Omni", trust_remote_code=True, device_map="auto") - Notebooks
- Google Colab
- Kaggle
| { | |
| "_from_model_config": true, | |
| "bos_token_id": 1, | |
| "eos_token_id": [2, 11], | |
| "pad_token_id": 0, | |
| "do_sample": true, | |
| "temperature": 0.6, | |
| "top_p": 0.95, | |
| "max_new_tokens": 16384, | |
| "reasoning_budget": 16384, | |
| "reasoning_grace": 512, | |
| "repetition_penalty": 1.0, | |
| "transformers_version": "4.55.4" | |
| } | |