Home
LLMs
Nemotron 3 Ultra 550B A55B

Nemotron 3 Ultra 550B A55B

The largest of NVIDIA's Nemotron 3 line, a 550B LatentMoE hybrid activating 55B per token across Mamba-2, MoE and attention layers with Multi-Token Prediction, under the OpenMDW-1.1 license.

Params: 550B, 55B active
Context: 1M tokens
Cutoff: Sep 2025
Released: Jun 2026
License: OpenMDW-1.1

Visit Nvidia Compare vs other LLMs →

Hosted API price

Cheapest via Azure · per 1M tokens

USD

Input: $0.66 /1M
Output: $2.64 /1M
Hosted by: 3 providers

Hosted API pricing

How we price

Order

Providers are listed cheapest input rate first, then by output rate, a tie going to the creator's own listing. A provider that lists the model without a published rate sits at the end.

Prices

Base-tier, on-demand rates in USD per 1M tokens, converted at ECB reference rates where a provider publishes in another currency. Cached-input and batch rates show only where a provider publishes them. "Cheapest" marks the one provider with the lowest input and output pair; a tie wears no badge.

Cost

Input rate times the input tokens plus output rate times the output tokens, at the monthly volume set above the table. Cached-input, batch, long-context and reasoning-token billing are not modelled.

Transparency and funding

Affiliates: Affiliate links are marked. We may earn a commission if you click them, but commissions never affect the order.

Every provider serving Nemotron 3 Ultra 550B A55B, cheapest input first. Set your monthly volume to see what each would bill.

Input M Output M

Every provider serving Nemotron 3 Ultra 550B A55B, with its price per 1M input and output tokens and what the volume above would cost
Provider	Input /1M	Output /1M	Cached input /1M	Cost at 10M in + 2M out	Link
Azure Cheapest	$0.66	$2.64	$0.13	$11.88	Visit website
DigitalOcean	$0.90	$1.70		$12.40	Visit website
Nvidia Creator	Open weights, no hosted price				Visit website

Heads up: Base-tier, on-demand rates per 1M tokens; cached, batch and long-context tiers excluded. A provider may serve a shorter context or a quantized build than the creator's release. Verify before provisioning. More on how we price.

Estimated cost to self-host

Nemotron 3 Ultra 550B A55B needs about 313 GB of GPU memory at 4-bit with 32K context. The cheapest rental that fits is 8× A40 at about $2,822 a month.

That costs the same as roughly 3B tokens a month on Azure's API. Self-hosting is more expensive below that volume.

Single node ≤ 8

What it costs to self-host Nemotron 3 Ultra 550B A55B at each precision: the memory it needs, the cheapest rental that holds it, that rental's month and the API volume it pays off at
Precision	Memory	Cheapest fit (32K context)(Full 1M context)	Cost /mo	Break-even vs API
4-bit INT4 / FP4	313 GB555 GB	8× A40	$2,822$10,138	3B tokens /mo10B tokens /mo
8-bit FP8 / INT8	560 GB677 GB	8× A100	$10,138$12,730	10B tokens /mo13B tokens /mo
16-bit FP16 / BF16	1,110 GB1,352 GB	8× B200	$36,000$45,446	36B tokens /mo46B tokens /mo

Estimates based on median on-demand rates for Nvidia GPUs. Memory is weights plus KV cache for one request, using FP8 KV cache where supported and FP16/BF16 otherwise. KV cache assumed at 250 KB per token, a conservative default. Break-even assumes a 5:1 input-to-output ratio. No guarantee of runtime support, usable performance, or that a matching quantized build exists. How we estimate costs.

Similarly priced models

The models nearest Nemotron 3 Ultra 550B A55B by blended rate, each at its own cheapest provider.

Models priced nearest Nemotron 3 Ultra 550B A55B per 1M tokens, each at its cheapest provider
Model	Blended /1M	Input /1M	Output /1M	Context	Cutoff	vs Nemotron 3 Ultra 550B A55B
DeepSeek R1 Distill Llama 70B DeepSeek	$0.80	$0.80	$0.80	131K		−19%
GLM-4.7 Z.AI	$0.8667	$0.60	$2.20	205K		−12%
Gemini 3 Flash Preview Google Cloud	$0.9167	$0.50	$3.00	1M	Jan 2025	−7%
Qwen3.6-Plus Alibaba Cloud	$0.9167	$0.50	$3.00	1M		−7%
GLM-5.2 Z.AI	$0.95	$0.70	$2.20	1M		−4%
Nemotron 3 Ultra 550B A55B This model Nvidia	$0.99	$0.66	$2.64	1M	Sep 2025
DeepSeek R1 DeepSeek	$1.00	$0.70	$2.50	128K		+1%
DeepSeek R1 Turbo DeepSeek	$1.00	$0.70	$2.50	64K		+1%
Llama 3 70B Meta	$1.00	$0.65	$2.75	8K	Dec 2023	+1%
Qwen3-VL-235B-A22B Thinking Alibaba Cloud	$1.00	$0.40	$4.00	262K		+1%
Kimi K2.6 Moonshot AI	$1.0333	$0.60	$3.20	262K		+4%

USD per 1M tokens at each model's cheapest listed provider. Blended is the cost of 10M input plus 2M output tokens, spread over the 12M.

Packet·ai

B200 @ $3.75/h on demand

No contract, no spot interrupts, no cluster minimums

Full root SSH, web dashboard, CI/CD ready

Reserve Now

Capabilities

What Nemotron 3 Ultra 550B A55B accepts and can do, as published by Nvidia.

Text Accepts and generates natural-language text.
Reasoning Thinks before it answers, always on or as a switchable mode.
Function calling Connect to external tools, APIs, and systems.
Structured output Return responses in structured formats like JSON.

Common questions

What is Nemotron 3 Ultra 550B A55B good for?

Complex multi-step agents, analysis and RAG over a very long context, at the heavy end of the Nemotron 3 line. Open weights under OpenMDW-1.1, on a multi-GPU node.

When is Nemotron 3 Ultra 550B A55B not a good fit?

Cost-sensitive or modest self-hosted use: it sits at the top of NVIDIA's price range and takes a multi-GPU node. Text only, with no published maximum output.

What is the cheapest way to run Nemotron 3 Ultra 550B A55B?

Hosted, unless you push serious volume. Azure charges $0.66 in / $2.64 out per 1M tokens. The cheapest rental that fits is 8x A40 at $2,822 a month, which costs the same as about 3B tokens a month on that API.

Can I self-host Nemotron 3 Ultra 550B A55B?

Yes. The weights are OpenMDW-1.1 licensed. At 4-bit it needs about 313 GB of GPU memory, which starts at roughly $2,822 a month on the cheapest rental that fits.