All models
Text & reasoningStable
NVIDIANVIDIA Nemotron 3 Ultra
NVIDIA's largest Nemotron 3 reasoning model for complex agents, coding, planning, tools, RAG, and million-token analysis.
Plain-English overview
What NVIDIA Nemotron 3 Ultra actually is
Nemotron 3 Ultra is a 550-billion-parameter mixture-of-experts model that activates 55 billion parameters for each token. It is built for demanding agent and reasoning work and supports up to one million tokens of text context with a configurable reasoning budget.
NVIDIA offers a prototype API, NIM deployment, partner endpoints, and downloadable weights. That range lets teams trade convenience for infrastructure control, but it also means performance and cost are properties of the chosen deployment rather than the checkpoint alone.
Good fit for
- Complex long-running agents
- Large-context reasoning and RAG
- Organizations deploying on NVIDIA infrastructure
Category comparison
The facts that matter for text models
These are provider-published specifications, not Cody benchmark scores. Follow the linked sources for current limits and endpoint-specific exceptions.
- Context window
- Up to 1M tokensMaximum combined prompt and working context documented by the provider.
- Maximum output
- Not separately publishedProvider-published response limit, where available.
- Knowledge cutoff
- Pretraining through Sep 2025; post-training through May 2026Latest reliable knowledge date explicitly published by the model provider. Search and connected tools can retrieve newer information but do not change the model's built-in cutoff.
- Input price
- Free prototype or deployment costCurrent standard list price per million input tokens unless noted.
- Output price
- Free prototype or deployment costCurrent standard list price per million output tokens unless noted.
- Inputs
- TextMedia types accepted by the listed model endpoint.
- Tools & agents
- Tool calling, agents, configurable reasoning budgetSelected native tools and agent-building capabilities, not an exhaustive list.
Pricing & comparisons
Estimate your cost
Set your usage. Your estimate updates as you type.
Example: 6,000 input tokens for 10 pages, plus a 500-token summary. Page lengths vary; adjust the numbers below.
One run sends your input to the model once and receives an answer. Tokens are pieces of text: input is what you send, output is the answer you receive.
Advanced options
NVIDIA Nemotron 3 Ultra
NVIDIA
Estimated total (USD)
For the usage above · USD · API pricing, not a subscription
How this estimate works
Estimates exclude taxes, tools, cache storage/writes, free allowances and custom discounts. Image estimates cover output only, not prompt or reference-image charges. Quality modes differ by model. Unlisted settings are not treated as free.
API and provider access
Where to get NVIDIA Nemotron 3 Ultra
Availability
Regions and access stage
Available through a free NVIDIA API prototype, partner endpoints, NVIDIA NIM deployment, and downloadable weights.
NVIDIA lists global deployment for the model; actual hosted regions and self-host location depend on the endpoint or infrastructure selected.
Check live availabilityData and training
The route matters.
Downloaded weights can run in infrastructure the operator controls. NVIDIA's hosted prototype and partner endpoints are governed by their own trial, logging, retention, and deployment terms.
This is a concise reading of the cited provider material, not legal advice. A third-party gateway can have different storage, routing, training, and residency terms from the model maker's direct API.
Read the provider policyFrequently asked questions
NVIDIA Nemotron 3 Ultra FAQ
What is NVIDIA Nemotron 3 Ultra?
NVIDIA's largest Nemotron 3 reasoning model for complex agents, coding, planning, tools, RAG, and million-token analysis. Nemotron 3 Ultra is a 550-billion-parameter mixture-of-experts model that activates 55 billion parameters for each token. It is built for demanding agent and reasoning work and supports up to one million tokens of text context with a configurable reasoning budget.
When was NVIDIA Nemotron 3 Ultra released?
NVIDIA Nemotron 3 Ultra was released on June 4, 2026 according to the cited provider materials.
Where can I access NVIDIA Nemotron 3 Ultra?
Available through a free NVIDIA API prototype, partner endpoints, NVIDIA NIM deployment, and downloadable weights. The access routes listed in this guide are NVIDIA NIM and Hugging Face.
How much does NVIDIA Nemotron 3 Ultra cost?
Free NVIDIA prototype endpoint; production and self-host cost varies. NVIDIA offers a prototype NIM endpoint plus partner and downloadable deployment routes. Production hosting and GPU costs depend on the chosen route.
Where is NVIDIA Nemotron 3 Ultra available?
Available through a free NVIDIA API prototype, partner endpoints, NVIDIA NIM deployment, and downloadable weights. NVIDIA lists global deployment for the model; actual hosted regions and self-host location depend on the endpoint or infrastructure selected.
Is my NVIDIA Nemotron 3 Ultra API data used for training?
Downloaded weights can run in infrastructure the operator controls. NVIDIA's hosted prototype and partner endpoints are governed by their own trial, logging, retention, and deployment terms. The policy belongs to the provider route and account terms, so verify it again before production use.
Price comparison
USD per million tokens for the shown API routes at standard context length. Cache and long-context rates may differ. Models without matching reviewed prices are omitted; lower cost does not mean better quality.
- Mistral Small 4$0.15
Mistral AI
- Gemini 3.8 Flash$0.75
Google
- GPT-5.6 Sol$4.00
OpenAI
- GPT-6 Astra$10.00
OpenAI
- Claude Fable 5.1$10.00
Anthropic
- Mistral Small 4$0.60
Mistral AI
- Gemini 3.8 Flash$3.75
Google
- GPT-5.6 Sol$20.00
OpenAI
- GPT-6 Astra$50.00
OpenAI
- Claude Fable 5.1$50.00
Anthropic
Related comparisons
Head-to-head comparisons
NVIDIA Nemotron 3 Ultra vs GPT-6 Astra
OpenAI's most capable model for difficult end-to-end work, combining frontier reasoning with a million-token context window and a broad set of agent tools.
Open full comparisonNVIDIA Nemotron 3 Ultra vs GPT-5.6 Sol
OpenAI's flagship general model for difficult coding, analysis, and professional work, with a very large context window and a broad native tool set.
Open full comparisonNVIDIA Nemotron 3 Ultra vs Claude Fable 5.1
Anthropic's frontier model for ambitious, long-running agentic and coding work, with strong vision and enterprise marketplace availability.
Open full comparisonNVIDIA Nemotron 3 Ultra vs Gemini 3.8 Flash
Google's stable, high-efficiency multimodal model for agents, software work, and large mixed-media inputs at an introductory Flash-tier price.
Open full comparisonNVIDIA Nemotron 3 Ultra vs Grok 4.6
SpaceXAI's frontier text-and-image model for coding, agentic tasks, and knowledge work, with direct web, X, and code tools.
Open full comparisonNVIDIA Nemotron 3 Ultra vs Grok 4.20 Multi-Agent Beta
A specialist Grok model that sends several AI agents to investigate a difficult question in parallel, then combines their work into one researched answer.
Open full comparisonNVIDIA Nemotron 3 Ultra vs Inkling
Thinking Machines Lab's large open-weights model for customizable reasoning, coding, tools, vision, and audio workflows.
Open full comparisonNVIDIA Nemotron 3 Ultra vs Mistral Medium 3.5
Mistral's open-weight frontier model for demanding multimodal, coding, reasoning, and agent workflows, with a 256K context window.
Open full comparisonNVIDIA Nemotron 3 Ultra vs Mistral Small 4
Mistral's lower-cost open model that combines normal instruction following, reasoning, coding, vision, and agent tools in one endpoint.
Open full comparisonNVIDIA Nemotron 3 Ultra vs Qwen3.8 Max
Alibaba's frontier Qwen model for long-context reasoning, coding, tools, and understanding text, images, and video.
Open full comparisonNVIDIA Nemotron 3 Ultra vs MiniMax M3
MiniMax's million-token multimodal model for coding, agents, computer use, and long projects at a low direct API price.
Open full comparisonNVIDIA Nemotron 3 Ultra vs GLM-5.3
Z.ai's open-weight long-context reasoning model for coding, tools, and agentic work across self-hosted and managed routes.
Open full comparisonNVIDIA Nemotron 3 Ultra vs Kimi K3
Moonshot AI's frontier multimodal model for million-token coding, reasoning, knowledge work, and tool-driven agents.
Open full comparisonNVIDIA Nemotron 3 Ultra vs Falcon-H1R-7B
TII's compact open reasoning model for text tasks, long contexts, and self-hosted function-calling workflows.
Open full comparisonNVIDIA Nemotron 3 Ultra vs Falcon-H1-34B-Instruct
TII's 34B open instruction model for multilingual text, coding, and controlled self-hosted applications.
Open full comparisonNVIDIA Nemotron 3 Ultra vs Muse Spark 1.3
Meta's latest multimodal reasoning model for long-running agents, coding, tools, and complex user collaboration.
Open full comparisonNVIDIA Nemotron 3 Ultra vs Celeris-1 Magnus
Celeris' agent-focused diffusion language model for fast reasoning, tool loops, structured actions, and OpenAI-compatible integration.
Open full comparisonNVIDIA Nemotron 3 Ultra vs NVIDIA Nemotron 3.5 Lightning
NVIDIA's compact 30B mixture-of-experts model for efficient specialist agents and high-volume text workflows.
Open full comparisonNVIDIA Nemotron 3 Ultra vs Qwen3.8-Flash
Alibaba's efficient million-context multimodal model for fast agents, coding, document work, vision, video understanding, and tool use.
Open full comparisonNVIDIA Nemotron 3 Ultra vs DeepSeek-V4-Pro
DeepSeek's flagship million-context text model for difficult reasoning, coding, long-running agents, tools, and very large outputs.
Open full comparisonNVIDIA Nemotron 3 Ultra vs DeepSeek-V4-Flash
DeepSeek's lower-cost V4 model for high-volume reasoning, coding, agents, and million-token text workloads.
Open full comparison
