RLTasksby ulam.ai
Models / NVIDIA

NVIDIA / MODEL EVIDENCE

Llama 3.3 Nemotron Super

6 benchmarks · 12 observations · 6 sources

Scores at a glance

6 benchmarks · large dot: Llama 3.3 Nemotron Super; small dots: closest models where reported

MATH-500 ↗71.2 %
71.271.2

Llama 3.3 Nemotron Super (Nonthinking) · MATH 500 · Accuracy ↗

1 more reported result

91.4 % · Llama 3.3 Nemotron Super (Thinking) · MATH 500 · Accuracy · Vals AI · MATH 500 ↗

GPQA Diamond ↗47.73 %
47.7347.73

Llama 3.3 Nemotron Super (Nonthinking) · GPQA Diamond · Accuracy ↗

1 more reported result

62.37 % · Llama 3.3 Nemotron Super (Thinking) · GPQA Diamond · Accuracy · Vals AI · GPQA Diamond ↗

MMLU-Pro ↗70.78 %
70.7870.78

Llama 3.3 Nemotron Super (Nonthinking) · MMLU Pro · Accuracy ↗

1 more reported result

69.58 % · Llama 3.3 Nemotron Super (Thinking) · MMLU Pro · Accuracy · Vals AI · MMLU Pro ↗

Closest models by score

Similar positions on shared benchmark leaderboards. Smaller gaps are closer.

There are not enough shared scored settings to identify close comparisons reliably.

Filter comparisons to this model →

Sources

Every plotted observation links to its evaluation. Reported configurations, benchmark versions and task subsets can differ; similarity describes shared scores, not training lift.

  1. AIME — evaluation methodology and results ↗

    Vals AI · 2026

  2. GPQA Diamond — evaluation methodology and results ↗

    Vals AI · 2026

  3. LiveCodeBench — evaluation methodology and results ↗

    Vals AI · 2026

  4. MATH 500 — evaluation methodology and results ↗

    Vals AI · 2026

  5. MMLU Pro — evaluation methodology and results ↗

    Vals AI · 2026

  6. TaxEval — evaluation methodology and results ↗

    Vals AI · 2026

Names used in source tables

Llama 3.3 Nemotron Super (Nonthinking) · Llama 3.3 Nemotron Super (Thinking)