NVIDIA Nemotron 3 Ultra

NVIDIA Nemotron 3 Ultra is a 550-billion-parameter Mixture-of-Experts model designed to enhance the efficiency and speed of long-running agents in complex workflows. It combines advanced reasoning capabilities with high throughput, enabling agents to perform tasks faster and more cost-effectively.

NVIDIA Nemotron 3 Ultra is a 550-billion-parameter Mixture-of-Experts model designed to enhance the efficiency and speed of long-running agents in complex workflows. It combines advanced reasoning capabilities with high throughput, enabling agents to perform tasks faster and more cost-effectively.

• 01550-billion-parameter Mixture-of-Experts model
• 02Hybrid Mamba-Transformer layers for efficient long-context handling
• 03NVFP4 quantization for cross-architecture GPU deployment with up to 5x higher throughput
• 04LatentMoE for expert routing
• 05Multi-token prediction for improved generative speed in multi-turn tasks
• 06Multi-Teacher On-Policy Distillation for continuous improvement and domain specialization