NVIDIA model product
Nemotron 3 Super
Nemotron 3 Super is a 120B total / 12B active parameter hybrid Mamba-Attention Mixture-of-Experts model optimized for agentic reasoning, coding, planning, tool calling, and long-context analysis.
Updated Aug 17, 2026. Default version: Nemotron 3 Super (120B A12B)
LLMBoard Score44.3Nemotron 3 Super (120B A12B)
Coverage100%23 benchmark families
Context window262.1KTokens
Official input price$0.20Nvidia API
Nemotron 3 Super Capability Profile
This profile uses the model's current scored version. Arena ratings and prices are shown separately.
Nemotron 3 Super Benchmark Results
Benchmark scores for Nemotron 3 Super (120B A12B).
Sort by
Benchmark | Score | Rank | Participants | Percentile | Evidence | Evaluated |
|---|
| BenchmarkWMT24++ | Score86.7% | Rank01 | Participants23 | Percentile100.0% | EvidenceC | Evaluated |
| BenchmarkRULER | Score91.8% | Rank02 | Participants4 | Percentile66.7% | EvidenceC | Evaluated |
| BenchmarkBird-SQL (dev) | Score41.8% | Rank05 | Participants7 | Percentile33.3% | EvidenceC | Evaluated |
| BenchmarkArena-Hard v2 | Score73.9% | Rank06 | Participants16 | Percentile66.7% | EvidenceC | Evaluated |
| BenchmarkHMMT 2025 | Score94.7% | Rank08 | Participants33 | Percentile78.1% | EvidenceC | Evaluated |
| BenchmarkLiveCodeBench | Score81.2% | Rank09 | Participants75 | Percentile89.2% | EvidenceC | Evaluated |
| BenchmarkMMLU-ProX | Score79.4% | Rank12 | Participants32 | Percentile64.5% | EvidenceC | Evaluated |
| BenchmarkMulti-Challenge | Score55.2% | Rank12 | Participants29 | Percentile60.7% | EvidenceC | Evaluated |
| BenchmarkSciCode | Score42.0% | Rank12 | Participants21 | Percentile45.0% | EvidenceC | Evaluated |
| BenchmarkAA-LCR | Score58.3% | Rank14 | Participants18 | Percentile23.5% | EvidenceC | Evaluated |
| BenchmarkIFBench | Score72.6% | Rank16 | Participants34 | Percentile54.5% | EvidenceC | Evaluated |
| BenchmarkTau2 Airline | Score56.3% | Rank18 | Participants23 | Percentile22.7% | EvidenceC | Evaluated |
| BenchmarkTerminal-Bench | Score25.8% | Rank22 | Participants25 | Percentile12.5% | EvidenceC | Evaluated |
| BenchmarkTau2 Retail | Score62.8% | Rank24 | Participants26 | Percentile8.0% | EvidenceC | Evaluated |
| BenchmarkTau2 Telecom | Score64.4% | Rank29 | Participants35 | Percentile17.6% | EvidenceC | Evaluated |
| BenchmarkMMLU-Pro | Score83.7% | Rank32 | Participants134 | Percentile76.7% | EvidenceC | Evaluated |
| BenchmarkSWE-bench Multilingual | Score45.8% | Rank35 | Participants38 | Percentile8.1% | EvidenceC | Evaluated |
| BenchmarkAIME 2025 | Score90.2% | Rank47 | Participants115 | Percentile59.6% | EvidenceC | Evaluated |
| BenchmarkTerminal-Bench 2.0 | Score31.0% | Rank51 | Participants51 | Percentile0.0% | EvidenceC | Evaluated |
| BenchmarkBrowseComp | Score31.3% | Rank58 | Participants62 | Percentile6.6% | EvidenceC | Evaluated |
| BenchmarkHumanity's Last Exam | Score22.8% | Rank58 | Participants99 | Percentile41.8% | EvidenceC | Evaluated |
| BenchmarkGPQA | Score82.7% | Rank75 | Participants239 | Percentile68.9% | EvidenceC | Evaluated |
| BenchmarkSWE-Bench Verified | Score53.7% | Rank92 | Participants111 | Percentile17.3% | EvidenceC | Evaluated |
Nemotron 3 Super Arena Results
Preference and agent-evaluation results for the default version.
Sort by
Arena | Category | Rank | Rating / score | Votes | Observations | Result date |
|---|
| Arenatext factuality | Categoryoverall | Rank138 | Rating / score1389.5 | Votes7,509 | ObservationsN/A | Result date |
| Arenatext | Categoryoverall | Rank151 | Rating / score1377.5 | Votes7,535 | ObservationsN/A | Result date |
| Arenatext style control | Categoryoverall | Rank183 | Rating / score1360.3 | Votes7,535 | ObservationsN/A | Result date |
Nemotron 3 Super Pricing
Official vendor API pricing appears first, followed by individual provider offers.
- Official API
- $0.20 input, $0.80 output per 1M
- Official provider
- Nvidia
- Lowest third-party
- From $0.05 input, $0.25 output per 1M via NanoGPT
- Tracked offerings
- 13
Sort by
Provider | Provider model ID | Region | Input / 1M | Output / 1M | Context | Updated |
|---|
| ProviderKenari | Provider model IDnemotron-3-super-120b-a12b | Regionglobal | Input / 1MN/A | Output / 1MN/A | Context262.1K | Updated |
| ProviderNanoGPT | Provider model IDnvidia/nemotron-3-super-120b-a12b | Regionglobal | Input / 1M$0.05 | Output / 1M$0.25 | Context262.1K | Updated |
| ProviderOpenRouter | Provider model IDnvidia/nemotron-3-super-120b-a12b | Regionglobal | Input / 1M$0.085 | Output / 1M$0.40 | Context1M | Updated |
| ProviderEden AI | Provider model IDflexai/Nemotron-3-Super-120B-A12B | Regionglobal | Input / 1M$0.085 | Output / 1M$0.40 | Context262.1K | Updated |
| ProviderKilo Gateway | Provider model IDnvidia/nemotron-3-super-120b-a12b | Regionglobal | Input / 1M$0.085 | Output / 1M$0.40 | Context262.1K | Updated |
| ProviderPioneer | Provider model IDnvidia/NVIDIA-Nemotron-3-Super-120B-A12B-FP8 | Regionglobal | Input / 1M$0.09 | Output / 1M$0.45 | Context1M | Updated |
| ProviderVercel AI Gateway | Provider model IDnvidia/nemotron-3-super-120b-a12b | Regionglobal | Input / 1M$0.15 | Output / 1M$0.65 | Context256K | Updated |
| ProviderNvidia | Provider model IDnvidia/nemotron-3-super-120b-a12b | Regionglobal | Input / 1M$0.20 | Output / 1M$0.80 | Context262.1K | Updated |
| ProviderPerplexity Agent | Provider model IDnvidia/nemotron-3-super-120b-a12b | Regionglobal | Input / 1M$0.25 | Output / 1M$2.5 | Context1M | Updated |
| ProviderSynthetic | Provider model IDhf:nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-NVFP4 | Regionglobal | Input / 1M$0.30 | Output / 1M$1 | Context262.1K | Updated |
| ProviderNebius Token Factory | Provider model IDnvidia/nemotron-3-super-120b-a12b | Regionglobal | Input / 1M$0.30 | Output / 1M$0.90 | Context256K | Updated |
| ProviderEden AI | Provider model IDnebius/nvidia/nemotron-3-super-120b-a12b | Regionglobal | Input / 1M$0.30 | Output / 1M$0.90 | Context8K | Updated |
| ProviderTensorX | Provider model IDnvidia/nemotron-3-super-120b-a12b | Regionglobal | Input / 1M$0.30 | Output / 1M$0.90 | Context262.1K | Updated |
Official prices use only the vendor's configured official Provider and positive standard USD PAYG rates. Third-party offers remain explicitly labeled.
Nemotron 3 Super Runtime Performance
Provider-specific output speed and catalog latency for Nemotron 3 Super (120B A12B). Runtime does not affect the capability score.
No runtime data
Browse runtime rankingsNo provider-specific speed or latency record is linked to the default version yet.
Output Speed is generated output tokens received per second. Catalog latency is reported separately from observed provider TTFT.
Nemotron 3 Super Specifications
Technical details for the model's default version.
- Version
- Nemotron 3 Super (120B A12B)
- Released
- Mar 11, 2026
- Knowledge cutoff
- Jun 1, 2025
- Parameters
- 120B
- Context window
- 262.1K
- Max output
- 262.1K
- Inputs
- text
- Outputs
- text
- Open weights
- Yes
- License
- NVIDIA Open Model License Agreement
Nemotron 3 Super Versions
Available versions of this model. The score column identifies the version used in the overall ranking.
Sort by
Version | Released | LLMBoard | Parameters | Context | Max output | Open weights | License |
|---|
| VersionNemotron 3 Super (120B A12B) | Released | LLMBoard44.3 | Parameters120B | Context262.1K | Max output262.1K | Open weightsYes | LicenseNVIDIA Open Model License Agreement |
Nemotron 3 Super vs nearby models
Open a comparison with the three ranked models immediately above and below this model.
Models similar to Nemotron 3 Super
Recommendations prioritize the same model type and family, then the closest LLMBoard score.
NV
Nemotron 3.5 Lightning
NVIDIA
NV
Nemotron 3 Nano
NVIDIA
NV
Nemotron Nano 9B
NVIDIA
NV
Llama 3.1 Nemotron Ultra 253B
NVIDIA
NV
Nemotron 3 Ultra
NVIDIA
NV
Llama 3.3 Nemotron Super 49B
NVIDIA
What is Nemotron 3 Super?
Key information about Nemotron 3 Super and its available data.
Nemotron 3 Super is a 120B total / 12B active parameter hybrid Mamba-Attention Mixture-of-Experts model optimized for agentic reasoning, coding, planning, tool calling, and long-context analysis.
It introduces LatentMoE (projecting tokens into a compressed latent space for expert routing, enabling 4x more experts at the same inference cost), Multi-Token Prediction for native speculative decoding (up to 3x faster generation), and native NVFP4 pretraining on Blackwell.
The hybrid architecture interleaves Mamba-2 layers for linear-time sequence processing with strategically placed Transformer attention layers as global anchors, supporting a 1M-token context window. 2 million rollouts. 2x higher throughput than GPT-OSS-120B while maintaining comparable accuracy.
Data as of 2026-08-17.
FAQ
Common questions about Nemotron 3 Super.
When was Nemotron 3 Super released?
Nemotron 3 Super's default version was released on Mar 11, 2026.
How much does Nemotron 3 Super cost?
Nemotron 3 Super's official API price is $0.20 per million input tokens and $0.80 per million output tokens via Nvidia. The lowest tracked third-party offer starts at $0.05 input and $0.25 output via NanoGPT.
Who created Nemotron 3 Super?
Nemotron 3 Super was created by NVIDIA.
What is the context window for Nemotron 3 Super?
The default version has a 262.1K token context window.
Is Nemotron 3 Super open weight?
Yes. The default version is marked as open weight under NVIDIA Open Model License Agreement .
How many API providers offer Nemotron 3 Super?
13 provider offerings are linked to the default version.
What models should I compare Nemotron 3 Super with?
Nearby ranked alternatives include o4 mini, MiniMax M2, K EXAONE 236B A23B.