Back to News
NVIDIA•June 4, 2026
NVIDIA Nemotron 3 Ultra
Open Source
AI
Machine Learning
Open Source
NVIDIA
NVIDIA Nemotron 3 Ultra
Nemotron 3 Ultra
Nemotron 3
### TL;DR
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.
Key Insights & Metrics
Pricing
Free
Cost structure
Version
3 Ultra
Current release version
Hardware
Compatible with NVIDIA Hopper, Blackwell, and Ampere GPUs
Compute requirements
Category
Open Source
Licensing model
Region
USA
Primary region
Key Features
- 550-billion-parameter Mixture-of-Experts model
- Hybrid Mamba-Transformer layers for efficient long-context handling
- NVFP4 quantization for cross-architecture GPU deployment with up to 5x higher throughput
- LatentMoE for expert routing
- Multi-token prediction for improved generative speed in multi-turn tasks
- Multi-Teacher On-Policy Distillation for continuous improvement and domain specialization
Discussion
0
Upvotes
0
Downvotes
0 reviews
Sign in to leave a review
Reviews
No reviews yet. Be the first to review!