Task
Text Generation
N
nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4
Text Generation·nvidia· 483.6K· 233
:---:--- Total Parameters 550B (55B active) Architecture LatentMoE - Mamba-2 + MoE + Attention hybrid with Multi-Token Prediction (MTP) Context Length Up to 1M tokens Minimum GPU Requirement 4xGB200, 4xB200, 4x GB300, 4x B300, 8xH100 Supported Languages English, French, Spanish, Italian, German, Japanese, Hindi, Korean, Brazilian Portuguese, and Chinese Best For Frontier reasoning
# pull & run locally
pip install mlforge-sdk && mlforge pull nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4
pip install mlforge-sdk && mlforge pull nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4
Model details
Provider
nvidia
Framework
transformers
Parameters
335.0B
Size
656 GB
License
other
Downloads
483.6K
Likes
233
Updated
2026-06-24
About nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4
:---:--- Total Parameters 550B (55B active) Architecture LatentMoE - Mamba-2 + MoE + Attention hybrid with Multi-Token Prediction (MTP) Context Length Up to 1M tokens Minimum GPU Requirement 4xGB200, 4xB200, 4x GB300, 4x B300, 8xH100 Supported Languages English, French, Spanish, Italian, German, Japanese, Hindi, Korean, Brazilian Portuguese, and Chinese Best For Frontier reasoning, complex agentic workflows, long-context analysis, tool use, multilingual reasoning, high-stakes RAG Reasoning Mode Configurable on/off via chat template (enablethinking=True/False) License OpenMDW License Agreement, version 1.1 Release Date June 4, 2026