Nemotron 3.5 Lightning: benchmark results
Provider: NVIDIA. Released 2026-08-11. Access: Open.
Unified ELO 1581 ± 1, rank #317 of 1392 rated models, from 57 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| LLM Stats (PinchBench) | 85.37 | Score (%) | 80 |
| SpeechMap Compliance | 75 | % Requests Completed | 70.1 |
| AA Omniscience | -17.72 | Score | 66 |
| AI for Education Pedagogy - Social studies | 81.82 | Accuracy (%) | 63.9 |
| Artificial Analysis Intelligence Index | 16.38 | Intelligence Index | 60.8 |
| AI Chess Leaderboard (Reasoning) | 752 | Elo | 60.1 |
| AA Humanity's Last Exam | 10.57 | Accuracy (%) | 55.7 |
| AA GPQA Diamond | 74.34 | Accuracy (%) | 55.6 |
| AA Long Context Reasoning | 60.33 | Accuracy (%) | 55.5 |
| AI for Education Pedagogy - Science | 81.97 | Accuracy (%) | 54.8 |
| ZeroEval GPQA Diamond | 75.44 | GPQA Diamond Score | 54.2 |
| AI for Education SEND | 74.31 | Accuracy (%) | 52.4 |
Interactive version: theaggregate.ai/model?slug=nemotron-3-5-lightning · How It Works · Data refreshed daily, snapshot 2026-09-05.