NVIDIA · model
Nemotron 3.5 Lightning 30B A3B NVFP4 benchmark scores
As of 2026-10-07, Nemotron 3.5 Lightning 30B A3B NVFP4 (NVIDIA) has measured scores on 21 benchmarks and estimated scores on 80 more.
| Benchmark | Score | Source |
|---|---|---|
| AA-GPQA Diamond | 74.3% | measured |
| AA-HLE | 10.6% | measured |
| AA-SciCode | 32.1% | measured |
| Artificial Analysis Intelligence Index | 12.9% | measured |
| CritPt | 0.0% | measured |
| GDPval-AA | 7.0% | measured |
| GPQA Diamond | 75.6% | measured |
| HLE | 15.8% | estimated ± 5.7 pp, medium confidence |
| IFBench | 72.9% | measured |
| AA-LCR | 49.2% | measured |
| LiveCodeBench v6 | 74.3% | estimated ± 3.1 pp, high confidence |
| AA-Omniscience Accuracy | 14.4% | measured |
| SciCode | 31.4% | measured |
| Terminal-Bench 2.1 | 23.5% | measured |
| BrowseComp | 36.8% | measured |
| HLE w/ tools | 33.2% | estimated ± 4.6 pp, medium confidence |
| IFEval | 92.8% | estimated ± 3.5 pp, high confidence |
| LongBench v2 | 0.0% | estimated ± 5.9 pp, low confidence |
| VITA-Bench | 5.2% | estimated ± 9.9 pp, low confidence |
| τ²-bench results | 79.9% | estimated ± 13.3 pp, low confidence |
| AA Coding Index | 26.8% | measured |
| APEX-Agents | 7.0% | estimated ± 10.0 pp, low confidence |
| APEX-Agents-AA | 6.0% | estimated ± 4.6 pp, medium confidence |
| BioMysteryBench (human-difficult) | 21.4% | estimated ± 5.9 pp, low confidence |
| DeepSearchQA | 61.7% | estimated ± 4.5 pp, medium confidence |
| SWE-bench Verified | 52.8% | measured |
| AutomationBench | 28.4% | estimated ± 9.7 pp, low confidence |
| BFCL v4 | 65.5% | estimated ± 2.6 pp, high confidence |
| CyberGym | 44.8% | estimated ± 4.6 pp, medium confidence |
| JobBench | 0.0% | estimated ± 8.6 pp, low confidence |
| SWE-bench Pro | 27.2% | estimated ± 3.2 pp, high confidence |
| τ³-bench results | 9.5% | measured |
| WideResearch | 50.8% | estimated ± 4.6 pp, medium confidence |
| AA Agentic Index | 6.1% | measured |
| MMLU-Pro | 81.6% | measured |
| GPQA | 75.6% | measured |
| AA-IFBench | 58.4% | estimated ± 8.5 pp, medium confidence |
| Gert Labs | 20.9% | estimated ± 7.4 pp, low confidence |
| AA-AnalystAgent | 4.4% | estimated ± 12.0 pp, low confidence |
| AA EnterpriseOps-Gym | 32.5% | estimated ± 6.7 pp, low confidence |
| AA Harvey LAB | 79.4% | estimated ± 1.2 pp, low confidence |
| ARC-AGI-1 | 60.3% | estimated ± 2.7 pp, medium confidence |
| ARC-AGI-2 | 24.3% | estimated ± 10.1 pp, low confidence |
| CursorBench 3.1 | 0.0% | estimated ± 3.4 pp, low confidence |
| CursorBench 3.2 | 15.9% | estimated ± 4.4 pp, medium confidence |
| Terminal-Bench 3.0 | 1.1% | estimated ± 9.1 pp, low confidence |
| FrontierCode 1.1 Main | 0.1% | estimated ± 2.7 pp, low confidence |
| FrontierSWE v2 | 0.0% | estimated ± 14.6 pp, low confidence |
| OSWorld-Verified | 46.2% | estimated ± 5.4 pp, low confidence |
| Vals GPQA Diamond | 74.4% | estimated ± 2.0 pp, medium confidence |
| Vals LiveCodeBench | 81.7% | estimated ± 0.8 pp, low confidence |
| Vals MMLU-Pro | 50.2% | estimated ± 1.2 pp, medium confidence |
| Vals SWE-bench | 38.4% | estimated ± 5.3 pp, low confidence |
| Terminal-Bench 2.1 (Vals AI) | 24.1% | estimated ± 7.4 pp, low confidence |
| AA AutomationBench | 2.6% | estimated ± 5.7 pp, low confidence |
| GDP.pdf | 9.4% | estimated ± 5.4 pp, low confidence |
| AA Tau3 Banking | 22.0% | estimated ± 5.8 pp, low confidence |
| AA Terminal-Bench 2.1 | 52.1% | estimated ± 3.5 pp, medium confidence |
| AA Terminal-Bench 4.0 | 0.0% | estimated ± 8.9 pp, low confidence |
| CursorBench 4.0 | 6.3% | estimated ± 1.6 pp, medium confidence |
| CWE-bench v1 | 11.9% | estimated ± 6.9 pp, low confidence |
| DeepSWE | 0.4% | estimated ± 6.4 pp, low confidence |
| HLE w/o tools | 10.5% | measured |
| OSWorld 2.0 | 17.7% | estimated ± 13.8 pp, low confidence |
| PostTrainBench v1.1 | 8.4% | estimated ± 6.1 pp, low confidence |
| SWE Multilingual | 36.5% | measured |
| Toolathlon-Verified | 20.8% | estimated ± 2.3 pp, low confidence |
| Vibe Code Bench | 0.6% | estimated ± 8.1 pp, low confidence |
| ExploitGym | 0.0% | estimated ± 6.1 pp, low confidence |
| C-Eval | 84.5% | estimated ± 0.9 pp, low confidence |
| Claw-Eval | 62.4% | estimated ± 7.9 pp, low confidence |
| DeepPlanning | 13.0% | estimated ± 8.3 pp, low confidence |
| MCP Atlas | 64.6% | estimated ± 7.8 pp, low confidence |
| MCP-Tasks | 0.0% | estimated ± 3.6 pp, low confidence |
| MMLU-ProX | 68.6% | estimated ± 1.7 pp, low confidence |
| MMLU-Redux | 91.2% | estimated ± 0.7 pp, high confidence |
| NL2Repo | 18.0% | estimated ± 5.1 pp, low confidence |
| NOVA-63 | 35.7% | estimated ± 1.6 pp, low confidence |
| QwenClawBench | 55.9% | estimated ± 6.0 pp, low confidence |
| SuperGPQA | 55.2% | estimated ± 7.5 pp, medium confidence |
| Toolathlon | 3.0% | estimated ± 3.5 pp, medium confidence |
| HealthBench Hard | 0.0% | estimated ± 5.1 pp, low confidence |
| MedXpertQA (Text) | 27.7% | estimated ± 4.5 pp, low confidence |
| MMLU-Pro (Arcee) | 81.4% | estimated ± 3.4 pp, medium confidence |
| React Native Evals | 70.8% | estimated ± 3.5 pp, high confidence |
| ResearchClawBench | 9.0% | estimated ± 2.0 pp, medium confidence |
| SWE-Rebench | 57.0% | estimated ± 3.6 pp, low confidence |
| ARC-AGI-3 | 0.0% | estimated ± 2.1 pp, medium confidence |
| BioMysteryBench (human-solvable) | 81.7% | estimated ± 0.6 pp, low confidence |
| DRACO | 64.4% | estimated ± 3.0 pp, low confidence |
| FrontierCode 1.1 Extended | 26.3% | estimated ± 3.4 pp, low confidence |
| HealthBench (raw) | 15.9% | estimated ± 6.4 pp, low confidence |
| HealthBench (length-adjusted) | 53.9% | estimated ± 3.7 pp, low confidence |
| HealthBench Professional | 24.1% | estimated ± 4.8 pp, low confidence |
| HealthBench Professional (raw) | 29.1% | estimated ± 5.0 pp, low confidence |
| HLE-Verified | 15.3% | estimated ± 2.0 pp, low confidence |
| LABBench2 | 55.7% | estimated ± 1.8 pp, low confidence |
| LiveCodeBench | 45.2% | estimated ± 8.9 pp, low confidence |
| OpenHarmony Bench | 51.0% | estimated ± 2.4 pp, medium confidence |
| MMMLU | 80.9% | estimated ± 1.1 pp, low confidence |
| MMLU | 81.3% | estimated ± 1.1 pp, medium confidence |