NVIDIA · model
Nemotron 3 Super 100B benchmark scores
As of 2026-11-22, Nemotron 3 Super 100B (NVIDIA) has measured scores on 7 benchmarks and estimated scores on 37 more.
| Benchmark | Score | Source |
|---|---|---|
| AA-GPQA Diamond | 77.2% | estimated ± 6.2 pp, medium confidence |
| AA-HLE | 15.2% | estimated ± 5.7 pp, medium confidence |
| AA-SciCode | 38.9% | estimated ± 3.8 pp, high confidence |
| AIME26 | 94.3% | estimated ± 1.0 pp, medium confidence |
| Artificial Analysis Intelligence Index | 12.7% | estimated ± 7.4 pp, medium confidence |
| CritPt | 0.5% | estimated ± 4.0 pp, high confidence |
| GPQA Diamond | 79.4% | estimated ± 1.2 pp, high confidence |
| HLE | 18.3% | measured |
| AA-Omniscience Accuracy | 20.7% | estimated ± 8.5 pp, medium confidence |
| SciCode | 38.2% | estimated ± 3.6 pp, high confidence |
| AA Coding Index | 28.6% | estimated ± 5.3 pp, medium confidence |
| SWE-bench Verified | 60.5% | measured |
| SWE-bench Pro | 36.9% | estimated ± 4.5 pp, high confidence |
| MMLU-Pro | 83.7% | measured |
| AIME 2024 | 92.0% | estimated ± 2.6 pp, high confidence |
| BBH | 98.1% | estimated ± 2.8 pp, medium confidence |
| GPQA | 79.2% | measured |
| MMLU | 90.1% | estimated ± 2.1 pp, high confidence |
| Vals GPQA Diamond | 77.1% | estimated ± 4.5 pp, high confidence |
| Vals LiveCodeBench | 60.2% | estimated ± 10.0 pp, low confidence |
| Vals MMLU-Pro | 82.7% | estimated ± 2.3 pp, high confidence |
| MATH-500 | 98.3% | estimated ± 2.0 pp, high confidence |
| MMMLU | 86.7% | estimated ± 1.7 pp, high confidence |
| AIME 2025 | 90.2% | measured |
| AA MMLU-Pro | 86.0% | estimated ± 1.2 pp, low confidence |
| ARC-AGI-1 | 48.9% | estimated ± 6.7 pp, low confidence |
| ARC-AGI-2 | 0.0% | estimated ± 10.9 pp, low confidence |
| Vals SWE-bench | 49.9% | estimated ± 6.6 pp, low confidence |
| DeepSWE | 14.5% | estimated ± 12.7 pp, low confidence |
| HLE w/o tools | 13.4% | estimated ± 5.3 pp, medium confidence |
| Vibe Code Bench | 0.0% | estimated ± 8.4 pp, low confidence |
| C-Eval | 88.5% | estimated ± 0.9 pp, low confidence |
| Claw-Eval | 5.5% | measured |
| HMMT Feb 2025 | 93.7% | measured |
| HMMT Feb 2026 | 85.6% | estimated ± 1.2 pp, medium confidence |
| HMMT Nov 2025 | 91.9% | estimated ± 1.5 pp, high confidence |
| MMAnswerBench | 81.4% | estimated ± 2.0 pp, low confidence |
| MMLU-Redux | 92.5% | estimated ± 0.6 pp, high confidence |
| NL2Repo | 12.5% | estimated ± 6.2 pp, low confidence |
| SuperGPQA | 59.4% | estimated ± 7.2 pp, medium confidence |
| React Native Evals | 70.9% | estimated ± 2.7 pp, low confidence |
| SWE-Rebench | 57.0% | estimated ± 3.6 pp, low confidence |
| SimpleQA | 37.9% | estimated ± 11.7 pp, low confidence |
| LiveCodeBench | 66.3% | estimated ± 14.6 pp, low confidence |