Side-by-side benchmark comparison across agentic, coding, multimodal, knowledge, reasoning, and math workflows.
Nemotron 3 Super 100B
56
Winner · 2/8 categoriesSarvam 30B
48
2/8 categoriesNemotron 3 Super 100B· Sarvam 30B
Pick Nemotron 3 Super 100B if you want the stronger benchmark profile. Sarvam 30B only becomes the better choice if knowledge is the priority or you want the stronger reasoning-first profile.
Nemotron 3 Super 100B is clearly ahead on the aggregate, 56 to 48. The gap is large enough that you do not need to squint at the spreadsheet to see the difference.
Nemotron 3 Super 100B's sharpest advantage is in agentic, where it averages 56.6 against 35.5. The single biggest benchmark swing on the page is HumanEval, 57% to 92.1%. Sarvam 30B does hit back in knowledge, so the answer changes if that is the part of the workload you care about most.
Sarvam 30B is the reasoning model in the pair, while Nemotron 3 Super 100B is not. That usually helps on harder chain-of-thought-heavy tests, but it can also mean more latency and more token spend in real use. Nemotron 3 Super 100B gives you the larger context window at 1M, compared with 64K for Sarvam 30B.
BenchLM keeps the benchmark table and the operator tradeoffs on the same page so a better score does not hide a materially slower, pricier, or smaller-context model.
Runtime metrics show N/A when BenchLM does not have a sourced snapshot for that exact model. The scoring rules and freshness policy are documented on the methodology page.
| Benchmark | Nemotron 3 Super 100B | Sarvam 30B |
|---|---|---|
| AgenticNemotron 3 Super 100B wins | ||
| Terminal-Bench 2.0 | 56% | — |
| BrowseComp | 61% | 35.5% |
| OSWorld-Verified | 54% | — |
| CodingNemotron 3 Super 100B wins | ||
| HumanEval | 57% | 92.1% |
| SWE-bench Verified | 44% | 34% |
| LiveCodeBench | 38% | — |
| LiveCodeBench v6 | — | 70.0% |
| Multimodal & Grounded | ||
| MMMU-Pro | 55% | — |
| OfficeQA Pro | 67% | — |
| Reasoning | ||
| MuSR | 60% | — |
| BBH | 83% | — |
| LongBench v2 | 75% | — |
| MRCRv2 | 75% | — |
| gpqaDiamond | — | 66.5% |
| KnowledgeSarvam 30B wins | ||
| MMLU | 65% | 85.1% |
| GPQA | 64% | — |
| SuperGPQA | 62% | — |
| MMLU-Pro | 72% | 80% |
| HLE | 13% | — |
| FrontierScience | 63% | — |
| SimpleQA | 62% | — |
| Instruction Following | ||
| IFEval | 84% | — |
| Multilingual | ||
| MGSM | 84% | — |
| MMLU-ProX | 77% | — |
| MathematicsSarvam 30B wins | ||
| AIME 2023 | 65% | — |
| AIME 2024 | 67% | — |
| AIME 2025 | 66% | 80% |
| HMMT Feb 2023 | 61% | — |
| BRUMO 2025 | 64% | — |
| MATH-500 | 83% | 97% |
| HMMT Feb 2025 | — | 73.3% |
| HMMT Nov 2025 | — | 74.2% |
Nemotron 3 Super 100B is ahead overall, 56 to 48. The biggest single separator in this matchup is HumanEval, where the scores are 57% and 92.1%.
Sarvam 30B has the edge for knowledge tasks in this comparison, averaging 80 versus 53.4. Inside this category, MMLU is the benchmark that creates the most daylight between them.
Nemotron 3 Super 100B has the edge for coding in this comparison, averaging 40.3 versus 34. Inside this category, HumanEval is the benchmark that creates the most daylight between them.
Sarvam 30B has the edge for math in this comparison, averaging 86.5 versus 69.6. Inside this category, AIME 2025 is the benchmark that creates the most daylight between them.
Nemotron 3 Super 100B has the edge for agentic tasks in this comparison, averaging 56.6 versus 35.5. Inside this category, BrowseComp is the benchmark that creates the most daylight between them.
Get notified when new models drop, benchmark scores change, or the leaderboard shifts. One email per week.
Free. No spam. Unsubscribe anytime. We only store derived location metadata for consent routing.