← All models

Nemotron 3 Super 120B A12B (Reasoning)

open weights
NVIDIA · released 2026-03-11 · 2 offers

Output 145 tokens/sFirst token 1.2 sContext 1M tokens

Top 2 cheapest providers (Adjusted $/task)

The same list price can give a different adjusted $/task (caching, token efficiency) — click a price for its inputs.

Within the active global provider, residency and confidentiality filters.

#ProviderAdjusted $/task
1AWS Bedrock
2Nebius

Composite

2 of 7 inputs6 radar axes: DesignArena's two boards share one

31.6

includes −3.7 for its Benchmaxxing signal (from 35.3; why, switch off in Options)

Benchmaxxing signal +3.7, light Benchmaxxing tag, uncertain: The 80 % interval reaches below zero — treat this tag as uncertain · light report →The 80 % interval reaches below zero — treat this tag as uncertain.

AA CodingCoding Agent v1.4AA IntelligenceAA AgenticEpoch ECISoftware ECIDesignArenaAA Coding: percentile 43AA Intelligence: percentile 57AA Agentic: percentile 31

AA Coding 37.7Coding Agent v1.4 AA Intelligence 13.6AA Agentic 4.1Epoch ECI Software ECI DesignArena —/—

Radar: percentile among all models measured on each input; a gap means not measured.

Benchmark sheet

20 of 140 registered benchmark versions · bars show the percentile among all models measured on each benchmark.

Compare this model →

Agentic

  • APEX-Agents-AA31.8%

    APEX-Agents-AA published 2026-09-10 · Published board — Tests professional-service tasks that require agents to produce locally graded deliverables.

  • GDPval-AA v229644 Elo

    GDPval-AA v2 v2 · Published board — Tests professional knowledge-work deliverables across occupations using AA's Stirrup harness.

  • ITBench-AA31.1%

    ITBench-AA published 2026-09-10 · Published board — Tests root-cause diagnosis from offline Kubernetes incident snapshots.

  • Terminal-Bench Hard (AA)7128.8%

    Terminal-Bench Hard (AA) pinned revision 74221fb · Published board — Tests a pinned 44-task hard subset of terminal-based work using Terminus 2.

  • Terminal-Bench v2.1 (AA)3938.6%

    Terminal-Bench v2.1 (AA) v2.1 · Published board — Tests terminal-based work on the 89-task verified refresh using Terminus 2.

  • Terminal-Bench v4.0 (AA)180.0%

    Terminal-Bench v4.0 (AA) v4.0 · Published board — Tests terminal-based work on the 66-task release using mini-SWE-agent v2.4.6.

  • AA Agentic Index314.10

    AA Agentic Index published 2026-09-18 · Published board — Artificial Analysis publishes one Agentic Index per model, measured on its primary configuration and relayed by OpenRouter's Benchmarks API; the value is attached at family scope on the deterministic representative. Artificial Analysis publishes this board without a version number; we keep the date each result was retained.

Coding

  • SciCode (AA subproblems) v1.0.11736.2%

    SciCode (AA subproblems) v1.0.1 · Published board — Tests scientific Python programming with scientist-annotated background information.

  • AA Coding Index4337.7

    AA Coding Index published 2026-09-19 · Published board — Artificial Analysis publishes this board without a version number; we keep the date each result was retained.

Instruction-following

Knowledge

Long-context

  • GDP.pdf (AA)222.6%

    GDP.pdf (AA) published 2026-09-10 · Published board — Tests professional reasoning over long PDFs with AA document preparation and grading.

  • AA-LCR v1.16165.7%

    AA-LCR v1.1 v1.1 · Published board — Tests reasoning across multiple long documents with corrected answer keys and grading.

  • MLCR-AA143.3%

    MLCR-AA published 2026-09-10 · Published board — Tests medical-record synthesis and reasoning across long, fragmented case documents.

Reasoning

Science

Tool-use

  • AutomationBench-AA v1.0.6293.8%

    AutomationBench-AA v1.0.6 · Published board — Tests multi-app SaaS workflows through REST tools on a held-out AutomationBench split.

  • EnterpriseOps-Gym-AA223.5%

    EnterpriseOps-Gym-AA published 2026-09-10 · Published board — Tests enterprise workflows through MCP tools against resettable application environments.

  • τ²-Bench Telecom (AA)6167.8%

    τ²-Bench Telecom (AA) published 2026-09-10 · Published board — Tests dual-control telecom agents that coordinate tool use with a simulated user.

  • τ³-Banking (AA) v1.0.13110.3%

    τ³-Banking (AA) v1.0.1 · Published board — Tests banking support agents that retrieve policies and change account state through tools.

Unusual results

2 threshold-crossing signals flagged · 21 eligible benchmark families

How flags are calculated

Heuristic screen, not statistical significance: benchmark families are correlated and source uncertainty is unknown. Peer evidence needs ≥ 20 independently measured matched configurations from ≥ 10 distinct model families. A flag needs a directed population z-score of magnitude ≥ 1.5 and a gap of ≥ 1.5 from the leave-one-benchmark-family-out mean z in the same direction, over ≥ 5 other benchmark families.

  • unusually weak ITBench-AA published 2026-09-10
    Why
    Observed: 0.0113 fraction
    Peer mean 0.31758 · peer sd 0.13369 · n 33 · families 31
    Directed z -2.291 · baseline z -0.341 · gap -1.95 · profile n 20
    0.0113 fractionmeasuredobserved 2026-09-10artificialanalysis.ai
    Evidence
    Axis: ITBench-AA · published 2026-09-10 · Published board
    Exact value: 0.0112994350282486 fraction
    Observed: 2026-09-10T21:47:16.627Z · publication date: not recorded
    Observation id: aa:63872e9c-3377-4a6b-b477-7bba244c38e9:itBenchSre
  • unusually weak EnterpriseOps-Gym-AA published 2026-09-10
    Why
    Observed: 0.23545 fraction
    Peer mean 0.38988 · peer sd 0.07948 · n 48 · families 46
    Directed z -1.943 · baseline z -0.358 · gap -1.584 · profile n 20
    0.23545 fractionmeasuredobserved 2026-09-10artificialanalysis.ai
    Evidence
    Axis: EnterpriseOps-Gym-AA · published 2026-09-10 · Published board
    Exact value: 0.23545210384959714 fraction
    Observed: 2026-09-10T21:47:16.627Z · publication date: not recorded
    Observation id: aa:63872e9c-3377-4a6b-b477-7bba244c38e9:enterpriseOpsGym
Missing coverage · 123 benchmark versions

No result does not mean a zero, or that the model was never tested. Collection failures and disputed versions retain their distinct status.

Variants / reasoning settings

Artificial Analysis snapshot 2026-09-19 · Data: Artificial Analysis · Data: Epoch AI (CC BY)

VariantAA CodingAA Intelligence
Nemotron 3 Super 120B A12B (Reasoning)37.713.6
NVIDIA: Nemotron 3 Super
Token offers by platform · 2 offers (Adjusted $/task)

Click any underlined price to see how it is estimated and where each input comes from. How we calculate adjusted cost.

2 offers within the active global filters; “—” means the catalog is active but no public token price is available.

AWS Bedrock (1)

AWS Bedrock

Nebius (1)

Nebius
Nemotron 3 Super 120B A12B (Reasoning) — benchmarks & cost | Benchmark Heaven