OpenModelsHQ

NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16

NVIDIALLMMoECustom license

NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16 is an open-weight language model from NVIDIA with 561B parameters (mixture of experts). At 4-bit it needs about 399.5 GB of memory, more than any single device in our directory has.

Overview

Estimated memory at 4-bit
399.5GB

Includes room for an 8k context and runtime overhead. More context needs more memory.

Total parameters
561B
Mixture of experts, all weights counted
Context window
256K
262,144 tokens
Cheapest API, blended / 1M
$0.74
₹70 · DeepInfra · 3:1 input to output
All NVIDIA API prices

Key information

Lab
NVIDIA
Released
3 Jun 2026
Parameters
561B · MoE
Type
Language
Context
262,144 tokens
HF downloads, 30 days
1,033,648
View all

Will it run on your hardware?

How we estimate
Memory needed vs devices
NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16 needs about 399.5 GB at 4-bit, 705.9 GB at 8-bit and 1331.8 GB at 16-bit. At 4-bit it fits none of the reference devices. Dashed lines are usable memory on each device; a model fits when its estimate stays within 90% of it. Log scale. 8k context.
Memory needed vs devices: data
Estimate or deviceMemory
Q4399.5 GB
Q8705.9 GB
BF161331.8 GB
RTX 5060 Ti 16 GB16 GB usable
RTX 5090 32 GB32 GB usable
DGX Spark 128 GB108.8 GB usable
Mac Studio M5 Ultra 256 GB192 GB usable

Single-device memory estimate. Pick a device to update the On your device column.

36 GB usable memory

FormatMemory neededOn your deviceExample hardware
4-bit (Q4)≈ 399.5 GBToo big 1110%No single device in our directory has enough memory
8-bit (Q8)≈ 705.9 GBToo big 1961%No single device in our directory has enough memory
16-bit (BF16)≈ 1331.8 GBToo big 3699%No single device in our directory has enough memory

Fits means the estimate stays within 90% of usable memory. Tight is above 90% and up to 100%. Too big is above 100%. These are memory checks, not speed or compatibility tests.

Mac allowances are 66% of total RAM up to 36 GB, and 75% above. AI boxes allow 85% because their memory is shared with the operating system.

Where to run it

On a Mac

No trusted MLX build found yet.

Check your Mac

With GGUF

No trusted GGUF build found yet.

Find compatible hardware

Through an API

DeepInfra (fp4) is the cheapest provider we found.

$0.40 / $1.76

Input / output per 1M tokens
₹38 / ₹166 at ₹94.29 per $1

Blended: $0.74 (₹70) per 1M. Ranked by a 3:1 mix of input and output tokens.

Hosted API 4

USD and INR per million tokensAll NVIDIA API prices
Precision
DeepInfraCheapest$0.74₹70$0.40₹38$1.76₹166262,144FP4
KT$0.925₹87$0.50₹47$2.20₹207262,144Not listed
BaseTen$1.05₹99$0.60₹57$2.40₹226202,800FP4
Venice$1.25₹118$0.625₹59$3.125₹295256,000FP8
View on OpenRouterEndpoint sourceSorted by the blended price: 3 parts input to 1 part output. Context limits differ by provider.

INR at ₹94.29 per $1. Prices fetched 14 Nov 2026.

More about this model

Frequently asked questions

How much memory does NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16 need? At an 8k context, about 399.5 GB at 4-bit, 705.9 GB at 8-bit and 1331.8 GB at 16-bit (BF16). A longer context needs more.

Can I run NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16 on a Mac? No Mac in our directory has enough usable memory for the 4-bit estimate of 399.5 GB. We have not found a trusted MLX or GGUF build yet.

Does the full context fit? NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16 supports up to 262,144 tokens of context. Our memory estimates use 8k, so long prompts need more memory than shown.

What does it cost through an API? The cheapest provider we found is DeepInfra, at $0.40 per 1M input tokens and $1.76 per 1M output tokens (₹38 and ₹166 at 94.29 rupees per dollar). Ranked by a 3:1 mix of input and output tokens.

License in plain English

Custom license. This model uses a custom license rather than a standard one. Read it on the model card before using the model commercially. This summary is not legal advice.

Read the license on the model card

OMHQ RigBench

Not yet tested. We show no speed or quality score without a measurement.

Revision history and sources

Last updated . Sources:Hugging Face model card, OpenRouter.