NVIDIA model product
3 Nemotron Super 49B is an NVIDIA model for reasoning, coding, and specialized AI agents.
Updated Sep 8, 2026. Default version: Llama 3.3 Nemotron Super 49B v1.5
This profile uses the model's current scored version. Arena ratings and prices are shown separately.
Benchmark scores for Llama-3.3 Nemotron Super 49B v1.
Benchmark | Score | Rank | Participants | Percentile | Evidence | Evaluated |
|---|
| BenchmarkMBPP | Score91.30% | Rank02 | Participants33 | Percentile96.88% | EvidenceC | Evaluated |
| BenchmarkMT-Bench | Score91.70% | Rank02 | Participants12 | Percentile90.91% | EvidenceC | Evaluated |
| BenchmarkBFCL v2 | Score73.70% | Rank03 | Participants5 | Percentile50.00% | EvidenceC | Evaluated |
| BenchmarkArena Hard | Score88.30% | Rank04 | Participants26 | Percentile88.00% | EvidenceC | Evaluated |
| BenchmarkMATH-500 | Score96.60% | Rank12 | Participants32 | Percentile64.52% | EvidenceC | Evaluated |
| BenchmarkAIME 2025 | Score58.40% | Rank102 | Participants119 | Percentile14.41% | EvidenceC | Evaluated |
| BenchmarkGPQA | Score66.67% | Rank152 | Participants247 | Percentile38.62% | EvidenceC | Evaluated |
| BenchmarkLM Arena Text | Score1,308.15 rating | Rank158 | Participants210 | Percentile24.88% | EvidenceA | Evaluated |
| BenchmarkLM Arena Text Style Control | Score1,327.70 rating | Rank162 | Participants210 | Percentile22.97% | EvidenceA | Evaluated |
Preference and agent-evaluation results for the default version.
Arena | Category | Rank | Rating / score | Votes | Observations | Result date |
|---|
| Arenatext | Categoryjapanese | Rank107 | Rating / score1,301.39 | Votes416 | ObservationsN/A | Result date |
| Arenatext | Categoryindustry mathematical | Rank113 | Rating / score1,395.90 | Votes177 | ObservationsN/A | Result date |
| Arenatext | Categorymath | Rank116 | Rating / score1,390.23 | Votes194 | ObservationsN/A | Result date |
| Arenatext style control | Categoryindustry mathematical | Rank116 | Rating / score1,399.80 | Votes177 | ObservationsN/A | Result date |
| Arenatext style control | Categoryjapanese | Rank117 | Rating / score1,299.30 | Votes416 | ObservationsN/A | Result date |
| Arenatext style control | Categorymath | Rank121 | Rating / score1,385.84 | Votes194 | ObservationsN/A | Result date |
| Arenatext | Categorypolish | Rank125 | Rating / score1,322.79 | Votes397 | ObservationsN/A | Result date |
| Arenatext | Categoryindustry medicine and healthcare | Rank130 | Rating / score1,366.72 | Votes195 | ObservationsN/A | Result date |
| Arenatext | Categoryindustry life and physical and social science | Rank131 | Rating / score1,366.81 | Votes588 | ObservationsN/A | Result date |
| Arenatext style control | Categoryindustry medicine and healthcare | Rank135 | Rating / score1,381.18 | Votes195 | ObservationsN/A | Result date |
| Arenatext style control | Categorypolish | Rank136 | Rating / score1,315.69 | Votes397 | ObservationsN/A | Result date |
| Arenatext | Categoryrussian | Rank137 | Rating / score1,331.80 | Votes204 | ObservationsN/A | Result date |
| Arenatext style control | Categorycoding | Rank137 | Rating / score1,402.88 | Votes653 | ObservationsN/A | Result date |
| Arenatext style control | Categoryindustry life and physical and social science | Rank137 | Rating / score1,376.52 | Votes588 | ObservationsN/A | Result date |
| Arenatext | Categorycreative writing | Rank138 | Rating / score1,307.90 | Votes366 | ObservationsN/A | Result date |
| Arenatext style control | Categoryindustry software and it services | Rank138 | Rating / score1,387.76 | Votes1,179 | ObservationsN/A | Result date |
| Arenatext style control | Categoryexpert | Rank140 | Rating / score1,368.92 | Votes178 | ObservationsN/A | Result date |
| Arenatext | Categoryindustry legal and government | Rank141 | Rating / score1,344.67 | Votes232 | ObservationsN/A | Result date |
| Arenatext | Categoryenglish | Rank142 | Rating / score1,357.85 | Votes1,379 | ObservationsN/A | Result date |
| Arenatext | Categoryexclude ties | Rank142 | Rating / score1,292.22 | Votes2,262 | ObservationsN/A | Result date |
| Arenatext style control | Categoryrussian | Rank142 | Rating / score1,346.67 | Votes204 | ObservationsN/A | Result date |
| Arenatext | Categoryoverall | Rank143 | Rating / score1,337.86 | Votes3,316 | ObservationsN/A | Result date |
| Arenatext | Categoryindustry software and it services | Rank144 | Rating / score1,358.87 | Votes1,179 | ObservationsN/A | Result date |
| Arenatext | Categorynon english | Rank144 | Rating / score1,315.92 | Votes1,937 | ObservationsN/A | Result date |
| Arenatext | Categoryexpert | Rank145 | Rating / score1,327.33 | Votes178 | ObservationsN/A | Result date |
| Arenatext | Categoryhard prompts | Rank145 | Rating / score1,333.80 | Votes1,429 | ObservationsN/A | Result date |
| Arenatext style control | Categoryenglish | Rank145 | Rating / score1,365.82 | Votes1,379 | ObservationsN/A | Result date |
| Arenatext style control | Categoryhard prompts | Rank145 | Rating / score1,361.97 | Votes1,429 | ObservationsN/A | Result date |
| Arenatext | Categorycoding | Rank146 | Rating / score1,354.81 | Votes653 | ObservationsN/A | Result date |
| Arenatext | Categorymulti turn | Rank146 | Rating / score1,332.56 | Votes562 | ObservationsN/A | Result date |
Official vendor API pricing appears first, followed by individual provider offers.
Provider | Provider model ID | Region | Input / 1M | Output / 1M | Context | Updated |
|---|
| ProviderNvidia | Provider model IDnvidia/llama-3.3-nemotron-super-49b-v1.5 | Regionglobal | Input / 1MN/A | Output / 1MN/A | Context131.1K | Updated |
| ProviderDeep Infra | Provider model IDnvidia/Llama-3.3-Nemotron-Super-49B-v1.5 | Regionglobal | Input / 1M$0.40 | Output / 1M$0.40 | Context131.1K | Updated |
Official prices use only the vendor's configured official Provider and positive standard USD PAYG rates. Third-party offers remain explicitly labeled.
Provider-specific output speed and catalog latency for Llama 3.3 Nemotron Super 49B v1.5. Runtime does not affect the capability score.
No 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.
Technical details for the model's default version.
Available versions of this model. The score column identifies the version used in the overall ranking.
Version | Released | LLMBoard | Parameters | Context | Max output | Open weights | License |
|---|
| VersionLlama 3.3 Nemotron Super 49B v1.5 | Released | LLMBoardN/A | ParametersN/A | Context131.1K | Max output131.1K | Open weightsYes | LicenseUnspecified |
| VersionLlama-3.3 Nemotron Super 49B v1 | Released | LLMBoard21.53 | Parameters49.9B | ContextN/A | Max outputN/A | Open weightsYes | LicenseLlama 3.1 Community License |
Recommendations prioritize the same model type and family, then the closest LLMBoard score.
Key information about Llama 3.3 Nemotron Super 49B and its available data.
3 Nemotron Super 49B is an NVIDIA Nemotron large language model (LLM) designed for reasoning, coding, and specialized AI agents.
Data as of 2026-09-08.
Common questions about Llama 3.3 Nemotron Super 49B.
Llama 3.3 Nemotron Super 49B's default version was released on Jul 25, 2025.
No official standard PAYG price is currently available for Llama 3.3 Nemotron Super 49B. The lowest tracked third-party offer starts at $0.40 input and $0.40 output via Deep Infra.
Llama 3.3 Nemotron Super 49B was created by NVIDIA.
The default version has a 131.1K token context window.
Yes. The default version is marked as open weight under an unspecified license.
2 provider offerings are linked to the default version.