indianboy712/nemotron-3-nano-omni-demo

NVIDIA Nemotron-3-Nano-Omni-30B-A3B-Reasoning-BF16 Demo

An interactive demonstration of NVIDIA Nemotron-3-Nano-Omni (30B-A3B-Reasoning), an open-source native multimodal large language model.

Key Features

• —Architecture: Mamba2-Transformer hybrid Mixture-of-Experts (MoE).
• —Parameters: 30B total parameters, 3B active parameters per forward pass (30B-A3B).
• —Context Window: 256k tokens.
• —Multimodal: Native processing of Text, Vision/Images, Video, and Audio.
• —Reasoning: Step-by-Step Chain-of-Thought (CoT) reasoning for complex logic and agentic workflows.