NVIDIA Nemotron models: open weights meant for your own GPUs
NVIDIA
Services found: 6
| # | AI model | Capabilities | API $/1M (in / out) | Context window | Price / mo | Output $/1M | Overview | Visit site |
|---|---|---|---|---|---|---|---|---|
| 1 |  Nemotron 3 Ultra NVIDIA |    | $0.625 / $3.125 | 262.1K tokens | API only | 3.125 | View | |
| 2 | Nemotron 3.5 Lightning NVIDIA | | $0.08 / $0.2 | 262.1K tokens | API only | 0.2 | View | |
| 3 | Nemotron 3 Super NVIDIA | | $0.085 / $0.4 | 262.1K tokens | API only | 0.4 | View | |
| 4 | Nemotron 3 Nano 30B A3B NVIDIA | | $0.05 / $0.2 | 262.1K tokens | API only | 0.2 | View | |
| 5 | Nemotron 3 Nano Omni NVIDIA |   | $0 / $0 | 256K tokens | API only | 0 | View | |
| 6 | Nemotron 3.5 Content Safety NVIDIA | | $0.2 / $0.2 | 131.1K tokens | API only | 0.2 | View |
Browse by need
NVIDIA ships Nemotron as open-weight models designed to run efficiently on its own accelerators. The line covers general instruction following and reasoning, with hosted access available through build.nvidia.com and self-hosting as the intended destination.
The argument is vertical integration: if your inference already runs on NVIDIA hardware, a model tuned for that stack removes a layer of optimisation work. For teams on managed APIs the appeal is weaker.
Hardware-alignedTuned for the accelerators you are probably already running.
Open weightsSelf-hosting is the intended path, not a fallback.
Enterprise focusBuilt around deployment rather than consumer chat.
Models from this vendor
6 NVIDIA models in this rating. Prices are input and output per 1M tokens from the vendor official price list; for dirhams multiply by 3.6725. Tap a card to open the model page.
Nemotron 3 Ultra
#1
NVIDIA
in $0.625out $3.125262.1K tokensText & chatReasoning
Nemotron 3.5 Lightning
#2
NVIDIA
in $0.08out $0.2262.1K tokensText & chatReasoning
Nemotron 3 Super
#3
NVIDIA
in $0.085out $0.4262.1K tokensText & chatReasoning
Nemotron 3 Nano 30B A3B
#4
NVIDIA
in $0.05out $0.2262.1K tokensText & chatReasoning
Nemotron 3 Nano Omni
#5
NVIDIA
in $0out $0256K tokensText & chatVision
Nemotron 3.5 Content Safety
#6
NVIDIA
in $0.2out $0.2131.1K tokensText & chatVision
How to choose
- Benchmark against a comparable model from a larger vendor before committing production traffic — line-ups this size get less independent testing.
- Start with Nemotron 3 Nano Omni at $0 / $0 per 1M tokens; it is the cheapest entry point here.
- Move up to Nemotron 3 Ultra at $0.625 / $3.125 only where the cheaper variant visibly fails, not by default.
- Check the context window against your longest input — the largest here is Nemotron 3 Ultra at 262.1K tokens.
Common mistakes
Assuming parity with the majorsOn the hardest tasks the frontier labs remain ahead; compare on your own cases.
Skipping the exit planA smaller vendor can change terms or availability faster. Identify a fallback model before you depend on this one.
Who this suits
Cost-driven teamsNemotron 3 Nano Omni at $0 / $0 undercuts most mainstream options.
EvaluatorsA useful comparison point when benchmarking the models against larger line-ups.
Regional usersRelevant where this vendor platform or language coverage matters more than raw capability.
Related filters
Text & chat228 models for text generation and chat
Reasoning169 models that think before answering
Tool calling166 models that call your functions
Frontier tier41 flagship models, the best each vendor sells
Alibaba (Qwen)34 models in this rating
OpenAI30 models in this rating
Google DeepMind17 models in this rating
Frequently asked questions
Do I need NVIDIA hardware?
Not strictly, but the efficiency argument assumes it.
Are Nemotron models free?
The weights are published; you pay for compute rather than tokens.
How do they compare with Llama-class models?
Comparable on general tasks, with tuning aimed at deployment efficiency.
How many NVIDIA models are in this rating?
6, covering text & chat, reasoning, tool calling.
Is NVIDIA a good default choice?
For most teams no — start with a mainstream vendor and use this one where its pricing or specialisation gives a concrete advantage.
Frequently searched
By capabilityReasoningCodingVisionLong context 1M+Free tierOpen weightsLow latencyRealtime voiceImage generation
By vendorAnthropic (13)OpenAI (30)Google DeepMind (17)xAI (8)Meta (6)DeepSeek (9)Mistral AI (11)Alibaba (Qwen) (34)Z.ai (12)Moonshot AI (5)
Rate this page
Help us improve