Can I run Nemotron Ultra locally?

No. NVIDIA has not released open weights for this model, so it is API-only and cannot be downloaded or self-hosted.

What does it cost and how large is the context window?

Pricing is $0.50 per million input tokens and $2.20 per million output tokens, with a published context window of 262,144 tokens; NVIDIA also claims support up to 1M tokens.

How does this differ from a standard dense frontier model?

Nemotron Ultra uses a mixture-of-experts architecture, which NVIDIA says delivers up to 5x faster inference and up to 30% lower cost specifically for agentic workloads compared to dense models at a similar capability level.

NVIDIA Nemotron 3 Ultra 550B A55B

6413°

#llm #nvidia #reasoning #agents #coding #research

Open DeepInfra

nvidia | 262,144 token context | $0.50/M input, $2.20/M output. Nemotron 3 Ultra is built for, frontier reasoning, orchestration, coding agents, deep research, and complex enterprise workflows. It delivers up to 5x faster inference and up to 30% lower cost for agentic workloads while supporting up to 1M token context.

Guides for this model

Open weights vs open source vs API-only: what you are actually allowed to do

How to choose an AI model in October 2026: the closed source news, then the open weights worth downloading

What is a GGUF file? Quantized models explained (Q4, Q8, fp8, NF4)

All guides

Captain's notes

Written by Captain, the house editor

Notes from 2026-10-13

Nemotron Ultra is NVIDIA's frontier mixture-of-experts model aimed at agentic and enterprise workloads: orchestration, deep research, complex coding pipelines. NVIDIA claims up to 5x faster inference and up to 30% lower cost compared to equivalent agentic tasks, which matters when you are running multi-step agent loops rather than single queries. It is a closed, API-only model, so you are renting compute, not downloading weights.

Known for

Built explicitly for multi-step agentic workflows and orchestration tasks.
Frontier reasoning depth suited to complex research and enterprise pipelines.
262,144-token context window with claimed support up to 1M tokens.
Competitive output pricing at $2.20/M tokens for a model at this capability tier.
Mixture-of-experts architecture delivering faster inference than dense equivalents.

Watch out for

Closed weights, API-only: no local deployment, no fine-tuning on your own hardware.
Output tokens at $2.20/M add up fast in long agentic loops that generate verbose reasoning traces.
NVIDIA has not published a model card here, so independent benchmark verification is thin.

Best for

Prompts to try

Multi-step research agent

You are a research agent. A fintech startup wants to enter the Brazilian market. Break the problem into sub-tasks, identify what information you would need to gather at each step, flag the key regulatory and competitive risks, and produce a structured plan of action with dependencies between steps clearly marked.

Complex code refactor plan

I have a Python monolith of roughly 40,000 lines that mixes database access, business logic and API handlers in the same files. Reason through a phased refactoring strategy: identify the safest extraction order, flag the highest-risk coupling points, and write the concrete steps for phase one including the test harness I should put in place first.

Deep reasoning: constraint puzzle

A logistics company has 6 warehouses, 4 truck types with different capacities and fuel costs, 12 delivery zones with time windows, and a rule that no driver works more than 9 hours. Walk through the reasoning to determine the minimum number of trucks needed for a given day, show your constraint analysis step by step, and identify which constraints are binding.

Compare with

Nemotron 3 Ultra
Open-weights model
262,144 token context · $0.60/M input, $2.40/M output
Nemotron 3.5 Lightning
Open-weights model
262,144 token context · $0.07/M input, $0.20/M output
Nemotron 3 Super
Open-weights model
262,144 token context · $0.08/M input, $0.45/M output
NVIDIA Nemotron 3 Super 120B A12B
Open-weights model
262,144 token context · $0.09/M input, $0.40/M output

More text models from nvidia

Questions people ask

Can I run Nemotron Ultra locally?: No. NVIDIA has not released open weights for this model, so it is API-only and cannot be downloaded or self-hosted.
What does it cost and how large is the context window?: Pricing is $0.50 per million input tokens and $2.20 per million output tokens, with a published context window of 262,144 tokens; NVIDIA also claims support up to 1M tokens.
How does this differ from a standard dense frontier model?: Nemotron Ultra uses a mixture-of-experts architecture, which NVIDIA says delivers up to 5x faster inference and up to 30% lower cost specifically for agentic workloads compared to dense models at a similar capability level.

Captain's verdict
A serious agentic workhorse from NVIDIA with genuinely competitive pricing for long multi-step pipelines, held back only by its closed, API-only nature.

Captain writes these notes from the model's own description and model card. Prices and context sizes are the provider's at the time of writing and can change.

Ranking badge

Updated Oct 22, 2026

Made this model? Show this badge on your page.

Paste it into your Hugging Face card or Civitai description.

Change colors / personalize / get code

Also ranked: #82 Hottest Text / LLM text model

Not this creator?

Get your own badge

How this score adds up

This model has no popularity data from any source yet, so it carries a stand-in number below 7,000° that keeps the lists complete. It never ranks above a model with data. The breakdown appears here as soon as a source reports it.

Glossary: the terms used here

Heat score: One number for how much a model is used and wanted right now, from 0° to 9,999°. Every list on this site that says hottest is sorted by it and nothing else.
Floor: The 7,000° every model with real data starts from. The scale runs from 0° to 9,999°; the part below the floor is reserved for stand-ins (models without data), so that a model with data always ranks above one without. The floor is the same for everyone and decides nothing; the degrees above it are the measurement.
Download-equivalents: Everything the score treats as a download, added up: downloads themselves, each torrent seeder as 1,000, each AI Horde image and each run on a generation site as one.
Log scale: Counting by tens, not by ones: going from 1,000 to 10,000 downloads adds the same degrees as going from 10,000 to 100,000. It keeps giants from crowding out everything else.
Likes: Thumbs up and likes from every source, worth half a download each on the log scale.
Adoption rate: How many downloads a day a model gets right now, from the difference between two readings of a source. A quarter of the weight of the download term.
Measured and estimated: Measured means two readings of a source exist and the difference is real. Estimated means only today's total is known and the past is spread over the model's life; the site says which one it is showing.
Seeder: Someone who keeps a copy of the weights online for others to download over a torrent, or a hosted mirror doing the same. Each one counts as 1,000 downloads.
AI Horde: A volunteer network that generates images for free on donated computers. The number of images it has made with a model is public, and each counts as one download.
Generation sites: Sites that host their own copies of a model and let people generate with it, such as PixAI, SeaArt and Yodayo. Their downloads count like Civitai's and each generation run as one download.
OpenRouter tokens: A text model served through an API, such as Claude or GPT, has no weights to download, so its usage is measured in tokens: the pieces of text it read and wrote through OpenRouter, a gateway many apps use, over the last 30 days. Every 10,000,000 tokens count as one download. Source: OpenRouter (openrouter.ai/rankings), CC BY 4.0.
Full model (checkpoint): The weights themselves, the file that generates on its own. It gets a bonus because add-ons depend on it.
LoRA and embedding: Small add-on files that steer someone else's weights toward a style, a character or a concept. They cannot run alone.
Family and weights: The base a model is built on, such as SD 1.5, SDXL, Flux or Wan. Every model inherits the release date of its family's weights.
Age: Degrees taken off for how long ago the family's weights came out: 150° a year, up to 750°. It keeps old downloads that had years to pile up from freezing the top of the list.
Family boost: A small, bounded nudge for LoRAs of a family this site is pushing forward right now, always listed in the breakdown.
Stand-in score: A stable number below the floor given to a model no source reports on yet, so lists stay complete. It never ranks above a model with data.
New! (grace period): A model less than 7 days old shows New! instead of a score and holds no place in the rankings. Its score is computed from day one, but a model one day old grows faster than anything, so without the wait the week's #1 would always be whatever came out yesterday rather than what held up for a week.
Trend: How a model's growth over a period compares with the typical model's growth over the same period. Faster than typical reads as a rise, slower as a fall.
Typical model: The middle of the catalog: half the models with a figure grew faster over the period, half slower.

How to install

How to Install Ollama and Run a Local AI (Windows, Mac, Linux)

Similar Models

Llama 3.2 1B Instruct

TEXT / LLM

AI Models for Image, Video & Text Generation

NVIDIA Nemotron 3 Ultra 550B A55B is a text model for the Text / LLM family listed on ModelTrends.ai, a read-only catalog of open source AI image, video and text models. Compare it with other Text / LLM models and check its heat score to see how it is trending.

More Text / LLM models llm nvidia reasoning agents coding research long context

Browse by Model Family