Model reference · open weights
NVIDIA-Nemotron-3-Ultra is an open-weight language model from NVIDIA. NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4 (NVFP4) weighs 352 GB; the smallest configuration that runs it is 8× A100 80 GB.
Summary of the nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4 model card, 2026-10-01
What it is
| Released by | NVIDIA |
|---|---|
| Released | 2026-06-03 |
| Parameters | 302.8B |
| VRAM | 352 GB for the weights |
What it runs on
How much memory each request adds isn't estimated yet for this architecture. The weights need at least the cards below, plus room for the context.
| Card | Weights alone |
|---|---|
| RTX 3060 12 GB … 8× H100 80 GB 13 smaller cards | does not fit |
| 8× A100 80 GB tensor parallel · FP4 without its speed-up here | fits |
| 8× H200 141 GB tensor parallel · FP4 without its speed-up here | fits |
| 8× B200 180 GB tensor parallel | fits |
Builds
| Build | Params | Precision | Weights | Smallest setup |
|---|---|---|---|---|
| NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4 (above) ↗ | 302.8B | NVFP4 | 352 GB | 8× A100 80 GB |
| NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16 ↗ | 560.5B | BF16 | 1121 GB | none of the reference configurations |
From the model card
| 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 (enable_thinking=True/False) |
| License | OpenMDW License Agreement, version 1.1 |
| Release Date | June 4, 2026 |
For more details on how to deploy and use the model - see the Quick Start Guide below!
Model Developer: NVIDIA Corporation
Model Dates: December 2025 - April 2026
Data Freshness:
NVIDIA Nemotron™ is a family of open models with open weights, training data, and recipes, delivering leading efficiency and accuracy for building specialized AI agents.
Nemotron-3-Ultra-550B-A55B-NVFP4 is a frontier-scale large language model (LLM) trained by NVIDIA, designed to deliver strong agentic, reasoning, and conversational capabilities. It is optimized for the most demanding workloads, including complex multi-step agents, long-context analysis, and high-accuracy reasoning over code, math, and science. Like other models in the family, it responds to user queries and tasks by first generating a reasoning trace and then concluding with a final response. The model's reasoning capabilities can be configured through a flag in the chat template.
The model employs a hybrid Latent Mixture-of-Experts (LatentMoE) architecture, utilizing interleaved Mamba-2 and MoE layers, along with select Attention layers. Like the Super model, the Ultra model incorporates Multi-Token Prediction (MTP) layers for faster text generation and improved quality, and it is trained using an NVFP4 pre-training recipe to maximize compute efficiency. The model has 55B active parameters and 550B parameters in total.
The supported languages include: English, French, Spanish, Italian, German, Japanese, Korean, Hindi, Brazilian Portuguese, and Chinese.
This model is ready for commercial and non-commercial use.
Governing Download Terms: Use of this model is governed by the OpenMDW-1.1 model license.
| Benchmark | Nemotron 3 Ultra BF16 | Nemotron 3 Ultra NVFP4 |
|---|---|---|
| Agentic | ||
| Terminal Bench 2.1 | 56.4 | 53.9 |
| GDPVal | 46.7 | 47.9 |
| SWE-Bench Verified | 70.7 | 69.5 |
| SWE-Bench Multilingual | 67.7 | 69.1 |
| ProfBench (Search) | 56 | 56.4 |
| PinchBench | 90 | 89.8 |
| TauBench V3 | ||
| Airline | 81.5 | 80.0 |
| Retail | 86.4 | 88.4 |
| Telecom | 92.9 | 93.6 |
| Banking | 22.6 | 19.2 |
| Average | 70.9 | 70.3 |
| BrowseComp | 44.4 | 41.4 |
| Reasoning and Knowledge | ||
| IOI 2025 | 570.0 | 564.7 |
| GPQA (no tools) | 87.0 | 87.9 |
| SciCode (subtask) | 44.6 | 43.5 |
| HLE (no tools) | 26.7 | 26.1 |
| CritPt (no tools) | 3.1 | 3.4 |
| OmniScience Accuracy | 24.1 | 24.6 |
| OmniScience Non-Hallucination | 78.7 | 75.5 |
| Chat & Instruction Following | ||
| IFBench (prompt) | 81.7 | 82.3 |
| Long Context | ||
| AA-LCR | 65.4 | 65.5 |
| RULER 1M | 94.7 | 94.0 |
All evaluation results were collected via Nemo Evaluator SDK. We used three main evaluation harnesses: Nemo Gym, Nemo Skills, and Harbor with extended sandboxing support via AWS ECS on Nemo Evaluator. In addition, the evaluations also used dedicated open-source packaged containers for ScaleAI Multi Challenge Multi Turn Instruction Following and KernelBench. For reproducibility purposes, more details on the evaluation settings and pinned containers can be found in the Nemo Evaluator SDK examples folder and the reproducibility tutorial for Nemotron 3 Ultra.
The following benchmarks are not onboarded yet in our open source tools and for these we used either their official open source implementation or otherwise an internal scaffolding that we plan to open source in the future: BrowseComp with Search, Tau Bench 3, ProfBench with Search, PinchBench, Vals.ai, LongBench v2.
NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4 is a frontier-scale general purpose reasoning and chat model intended to be used in English, Code, and supported multilingual contexts. This model is optimized for complex agentic workflows, long-context reasoning, and high-stakes analytical workloads. It is intended to be used by developers designing AI Agent systems, chatbots, RAG systems, and other AI-powered applications. This model is also suitable for complex instruction-following tasks and long-context reasoning over very large documents and codebases.
Hugging Face - 06/04/2026 via Hugging Face
Quoted from the model card on Hugging Face. The full card is behind the Hugging Face link above.