How to use nvidia/Nemotron-3-Nano-Omni-30B-A3B-Reasoning-FP8 with Transformers:
# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("nvidia/Nemotron-3-Nano-Omni-30B-A3B-Reasoning-FP8", trust_remote_code=True, dtype="auto")
