Nemotron 3 Super 120B A12B Benchmark Scores & Performance

Benchmark analysis of Nemotron 3 Super 120B A12B by NVIDIA across 32 sourced tests on BenchLM.

According to BenchLM.ai, Nemotron 3 Super 120B A12B ranks #58 out of 123 models with an overall score of 61/100. While not a frontier model, it offers specific advantages depending on the use case.

Nemotron 3 Super 120B A12B is a open weight model with a 256K token context window. It processes queries without explicit chain-of-thought reasoning, offering faster response times and lower token usage.

Its strongest category is Instruction Following (#37), while its weakest is Multimodal & Grounded (#75). This performance profile makes it a well-rounded choice across a range of tasks.

Creator

NVIDIA

Source Type

Open Weight

Reasoning

Non-Reasoning

Context Window

256K

Overall Score

61#58 of 123

Arena Elo

1278

Knowledge Benchmarks

MMLU
68
GPQA
67
SuperGPQA
65
OpenBookQA
63
MMLU-Pro
75
HLE
16
FrontierScience
66

Coding Benchmarks

HumanEval
59
SWE-bench Verified
46
LiveCodeBench
42
SWE-bench Pro
46

Mathematics Benchmarks

AIME 2023
67
AIME 2024
69
AIME 2025
68
HMMT Feb 2023
63
HMMT Feb 2024
65
HMMT Feb 2025
64
BRUMO 2025
66
MATH-500
85

Reasoning Benchmarks

SimpleQA
64
MuSR
62
BBH
88
LongBench v2
77
MRCRv2
77

Agentic Benchmarks

Terminal-Bench 2.0
55
BrowseComp
59
OSWorld-Verified
53

Multimodal & Grounded Benchmarks

MMMU-Pro
55
OfficeQA Pro
67

Instruction Following Benchmarks

IFEval
86

Multilingual Benchmarks

MGSM
86
MMLU-ProX
79

Frequently Asked Questions

How does Nemotron 3 Super 120B A12B perform overall in AI benchmarks?

Nemotron 3 Super 120B A12B ranks #58 out of 123 models with an overall score of 61. It is created by NVIDIA and features a 256K context window.

Is Nemotron 3 Super 120B A12B good for knowledge and understanding?

Nemotron 3 Super 120B A12B ranks #59 out of 123 models in knowledge and understanding benchmarks with an average score of 55.8. There are stronger options in this category.

Is Nemotron 3 Super 120B A12B good for coding and programming?

Nemotron 3 Super 120B A12B ranks #43 out of 123 models in coding and programming benchmarks with an average score of 44.2. There are stronger options in this category.

Is Nemotron 3 Super 120B A12B good for mathematics?

Nemotron 3 Super 120B A12B ranks #59 out of 123 models in mathematics benchmarks with an average score of 74.6. There are stronger options in this category.

Is Nemotron 3 Super 120B A12B good for reasoning and logic?

Nemotron 3 Super 120B A12B ranks #57 out of 123 models in reasoning and logic benchmarks with an average score of 71.8. There are stronger options in this category.

Is Nemotron 3 Super 120B A12B good for agentic tool use and computer tasks?

Nemotron 3 Super 120B A12B ranks #60 out of 123 models in agentic tool use and computer tasks benchmarks with an average score of 55.3. There are stronger options in this category.

Is Nemotron 3 Super 120B A12B good for multimodal and grounded tasks?

Nemotron 3 Super 120B A12B ranks #75 out of 123 models in multimodal and grounded tasks benchmarks with an average score of 60.4. There are stronger options in this category.

Is Nemotron 3 Super 120B A12B good for instruction following?

Nemotron 3 Super 120B A12B ranks #37 out of 123 models in instruction following benchmarks with an average score of 86. There are stronger options in this category.

Is Nemotron 3 Super 120B A12B good for multilingual tasks?

Nemotron 3 Super 120B A12B ranks #39 out of 123 models in multilingual tasks benchmarks with an average score of 81.5. There are stronger options in this category.

Is Nemotron 3 Super 120B A12B open source?

Yes, Nemotron 3 Super 120B A12B is an open weight model created by NVIDIA, meaning it can be downloaded and run locally or fine-tuned for specific use cases.

What is the context window size of Nemotron 3 Super 120B A12B?

Nemotron 3 Super 120B A12B has a context window of 256K, which determines how much text it can process in a single interaction.

Last updated: March 12, 2026

Weekly LLM Updates

New model releases, benchmark scores, and leaderboard changes. Every Friday.

Free. Your signup is stored with a derived country code for compliance routing.