NE

Nvidia-Nemotron-3-Ultra-550B-A55B

NemotronGGUF by unslothUpdated 2026-06-04
Download on Hugging Face
Parameters550B
Context1024K
Layers108
Architecturenemotron_h_moe
Smallest quant188.0 GB
Quick answer

Nvidia-Nemotron-3-Ultra-550B-A55B needs about 388.7 GB of memory at Q4_K_M (a 359.2 GB file) with an 8K context window, which is more than a 48 GB GPU holds. This repository has 22 quants, from 188.0 GB (IQ1_M) to 1099.0 GB (BF16). Longer context needs more: at 32K tokens the KV cache alone adds about 116.0 GB.

Nvidia Nemotron 3 Ultra 550B with 55B active parameters

Nvidia Nemotron 3 Ultra 550B A55B is a frontier‑scale LLM with a hybrid LatentMoE architecture. It supports up to 1,048,576 context tokens and is optimized for complex reasoning, long‑context analysis, and high‑accuracy code, math, and science tasks.

Good at

  • Long‑context reasoning
  • High‑accuracy code and math
  • Multi‑step agentic tasks

Limits

  • Requires large VRAM for full precision
  • No vision support
  • License governed by OpenMDW-1.1

Best for

  • Developing advanced AI agents
  • Research on long‑context LLMs
  • Complex problem solving

Who it’s for. Engineers and researchers building specialized AI agents who need strong reasoning and long‑context capabilities. The model is available for commercial and non‑commercial use under the OpenMDW license.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
IQ1_M188.0 GB (5 parts) 217.5 GB
Too big Get file →
IQ2_XXS193.8 GB (5 parts) 223.3 GB
Too big Get file →
IQ2_M194.0 GB (5 parts) 223.5 GB
Too big Get file →
Q2_K_XL202.2 GB (6 parts) 231.7 GB
Too big Get file →
IQ3_XXS223.3 GB (6 parts) 252.8 GB
Too big Get file →
IQ3_S250.5 GB (7 parts) 280.0 GB
Too big Get file →
Q3_K_M274.0 GB (7 parts) 303.5 GB
Too big Get file →
Q3_K_XL274.3 GB (7 parts) 303.8 GB
Too big Get file →
IQ4_XS286.1 GB (7 parts) 315.6 GB
Too big Get file →
IQ4_NL294.2 GB (8 parts) 323.7 GB
Too big Get file →
Q4_K_S327.7 GB (8 parts) 357.2 GB
Too big Get file →
MXFP4_MOE352.3 GB (9 parts) 381.8 GB
Too big Get file →
Q4_K_M359.2 GB (9 parts) 388.7 GB
Too big Get file →
Q4_K_XL360.2 GB (9 parts) 389.6 GB
Too big Get file →
Q5_K_S391.7 GB (10 parts) 421.2 GB
Too big Get file →
Q5_K_M426.5 GB (10 parts) 456.0 GB
Too big Get file →
Q5_K_XL427.2 GB (10 parts) 456.7 GB
Too big Get file →
Q6_K460.7 GB (11 parts) 490.2 GB
Too big Get file →
Q6_K_XL523.1 GB (12 parts) 552.6 GB
Too big Get file →
Q8_0584.3 GB (14 parts) 613.7 GB
Too big Get file →
Q8_K_XL594.5 GB (14 parts) 624.0 GB
Too big Get file →
BF161099.0 GB (24 parts) 1128.5 GB
Too big Get file →

For 8 GB VRAM, use IQ1_M or IQ2_XXS (≈217–223 GB memory at 8 k context). For 16 GB VRAM, Q4_K_M or Q4_K_XL (≈389 GB). For 24 GB VRAM, Q5_K_M or Q5_K_XL (≈456 GB). Memory estimates are approximate.

What GPU do you need for Nvidia-Nemotron-3-Ultra-550B-A55B?

These figures add the file size, an fp16 KV cache at an 8K context window (worked out from this model’s own header) and about 0.5 GB for the runtime. They are estimates, so leave a little headroom, and expect to use less context or a smaller quant if you are right at the limit.

  • 4 GB VRAM: no quant fits at an 8K context.
  • 6 GB VRAM: no quant fits at an 8K context.
  • 8 GB VRAM: no quant fits at an 8K context.
  • 12 GB VRAM: no quant fits at an 8K context.
  • 16 GB VRAM: no quant fits at an 8K context.
  • 24 GB VRAM: no quant fits at an 8K context.
  • 32 GB VRAM: no quant fits at an 8K context.
  • 48 GB VRAM: no quant fits at an 8K context.

Which Nvidia-Nemotron-3-Ultra-550B-A55B quant should you choose?

Q4_K_M is the usual starting point for Nvidia-Nemotron-3-Ultra-550B-A55B: it balances file size and quality well. Here it is a 359.2 GB file.

The Q8_0 file is 584.3 GB, about 1.6 times the size of Q4_K_M. It keeps the weights closer to the original, which only matters if you have the memory to spare.

If memory is tight, the smallest file is IQ1_M at 188.0 GB, with the quality trade-offs that come with fewer bits. Use it only when memory is very tight. Expect a clear drop in quality, which is usually worse on small models and easier to live with on large ones.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architecturenemotron_h_moe
Parameters550B (approximate, mixture of experts)
Active parameters55B
Layers108
Trained context1,048,576 tokens
KV cache (fp16)3456 KB per token

Source

Quantized byunsloth
FamilyNemotron
Licenseother
Downloads (HF)6,328
Last updated2026-06-04

Frequently asked questions about Nvidia-Nemotron-3-Ultra-550B-A55B

How much VRAM does Nvidia-Nemotron-3-Ultra-550B-A55B need?

About 388.7 GB at Q4_K_M with an 8K context window. That is the 359.2 GB file, plus the KV cache and roughly 0.5 GB of runtime overhead. Smaller quants need less and longer context needs more.

Can I run Nvidia-Nemotron-3-Ultra-550B-A55B on an 8 GB GPU?

Not at an 8K context. The smallest file needs about 217.5 GB, so you would need a larger GPU or CPU offloading.

What is the file size of Nvidia-Nemotron-3-Ultra-550B-A55B GGUF?

It depends on the quant: IQ1_M is 188.0 GB, IQ2_XXS is 193.8 GB, IQ2_M is 194.0 GB, Q2_K_XL is 202.2 GB, IQ3_XXS is 223.3 GB, IQ3_S is 250.5 GB and 16 more.

What context length does Nvidia-Nemotron-3-Ultra-550B-A55B support?

The GGUF header lists a trained context length of 1,048,576 tokens (1024K). The KV cache grows with the context you use: about 29.0 GB at 8K and 116.0 GB at 32K.

Where can I download Nvidia-Nemotron-3-Ultra-550B-A55B GGUF?

The files are on Hugging Face in the unsloth/NVIDIA-Nemotron-3-Ultra-550B-A55B-GGUF repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.

What license does Nvidia-Nemotron-3-Ultra-550B-A55B use?

Hugging Face lists the license as other. Licenses can differ between the quantized files and the original model, so check the repository and the base model before commercial use.

Can I run Nemotron 3 Ultra on 8 GB VRAM?

Use the IQ1_M or IQ2_XXS quantized files; they require about 217–223 GB of VRAM at 8 k context, which is beyond 8 GB, so it is not feasible.

Does this model support vision tasks?

No, the model has no vision capabilities.

What license applies to this model?

The model is governed by the OpenMDW License Agreement, version 1.1.

Other GGUFs of NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16

Similar models

Browse all →

More from unsloth

All by unsloth →

More Nemotron models

All Nemotron →

Other large models

All large models →