Nemotron 3 Ultra 550B-A55B
A solid pick for self-hosting and full data control.
Overview
Nemotron 3 Ultra 550B-A55B is NVIDIA's flagship open-weight model, a hybrid Mamba-2/MoE/Attention architecture with 550 billion parameters (55 billion active) built for complex reasoning, multi-step agentic workflows and document analysis up to one million tokens. It holds its own on scientific and coding benchmarks, though it trails rivals such as Kimi-K2.6 or DeepSeek-v4-Pro on several agentic tasks (SWE-Bench, BrowseComp). Released under the OpenMDW-1.1 licence with commercial use allowed, it can run on-premise or on European sovereign cloud, provided you have a serious server-grade GPU cluster (8x GB200/B200 or 16x H100).
Skill profile
Not disclosed
Three independent measures
What people prefer, what the model can do, what the market uses. Each source has its biases and none is enough on its own: we show them side by side, never blended.
Strengths
- World-class on Arena
- Very long context
- Open-source and self-hostable
Limitations
- API pricing not disclosed
Who is it for
- you want to control cost or self-host
- your data must stay in the EU
Ideal use cases
- Complex multi-step agents
- Long-document analysis
- High-stakes RAG
- Scientific and code reasoning
- On-premise sovereign deployment
Access & availability
Key specifications
What it costs per month
Estimate from sourced API prices and a hardware tier for self-hosting. Assumptions are editable.
Assumptions: 36-month amortisation, 0.25 €/kWh, 22 days/month, 1 $ = 0.92 €. Excludes labour, software licences, redundancy.
nAIvigate Trust Index
What the vendor offers and documents for professional use in Europe. An undocumented criterion counts as zero and stays visible: the score rises when the vendor publishes, not when we assume.
19/19 criteria documented · 100 % · 12 of them set by rule
See the criteria▾
Trust Index badge
Show this score in your docs, README or security page. It updates automatically.
[](https://naivigate.com/en/model/nemotron-3-ultra-550b)Data & compliance
What we know, line by line, with source and date. These lines feed the Data and Enterprise families of the Trust Index above. GDPR compliance depends on your processing, not on the model alone.
| EU data residency | Yes |
| Data processing agreement (DPA) | N/A· self-hosted: data never leaves |
| Transfer outside the EU | No |
| Training on your data | No |
| Configurable retention | N/A· self-hosted: data never leaves |
| Zero Data Retention | N/A· self-hosted: data never leaves |
| Self-hostable | Yes |
| Declared hosting | self |
Quality of this card’s data
This score measures the quality of the data on this card: price, context, Elo, licence, documented, sourced and fresh. It measures neither the model nor its compliance: for that, see the Trust Index above.
6/8 documented · 3/8 primary sources
Run it locally
| Quantization | Disk | RAM / VRAM | Typical hardware |
|---|---|---|---|
| Q4 · recommended | 313.5 GB | 347 GB | Server GPU infra (H100/A100…) |
| Q8 · balanced | 588.5 GB | 649 GB | Server GPU infra (H100/A100…) |
| FP16 · max quality | 1100 GB | 1212 GB | Server GPU infra (H100/A100…) |
Estimates for a moderate context. Long contexts need more RAM (KV cache).
Deploy
Copy-ready commands generated from this card. Adjust context length and GPU count to your hardware.
OpenAI-compatible server for production on NVIDIA GPUs.
pip install vllm
vllm serve nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16 \
--max-model-len 32768 \
--tensor-parallel-size 2 \
--dtype autoAdvanced data · for expertsArchitecture, modalities, detailed cost, full benchmarks▾
| Arena Elo | 1426 |
| MMLU | — |
| GPQA | — |
| HumanEval | — |
| SWE-Bench | — |
| MATH | — |