Side-by-side benchmark comparison across agentic, coding, multimodal, knowledge, reasoning, and math workflows.
Gemma 4 31B
73
Winner · 3/8 categoriesNemotron 3 Ultra 500B
60
1/8 categoriesGemma 4 31B· Nemotron 3 Ultra 500B
Pick Gemma 4 31B if you want the stronger benchmark profile. Nemotron 3 Ultra 500B only becomes the better choice if reasoning is the priority or you need the larger 10M context window.
Gemma 4 31B is clearly ahead on the aggregate, 73 to 60. The gap is large enough that you do not need to squint at the spreadsheet to see the difference.
Gemma 4 31B's sharpest advantage is in coding, where it averages 80 against 43.5. The single biggest benchmark swing on the page is LiveCodeBench, 80% to 41%. Nemotron 3 Ultra 500B does hit back in reasoning, so the answer changes if that is the part of the workload you care about most.
Nemotron 3 Ultra 500B gives you the larger context window at 10M, compared with 256K for Gemma 4 31B.
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 | Gemma 4 31B | Nemotron 3 Ultra 500B |
|---|---|---|
| Agentic | ||
| Terminal-Bench 2.0 | — | 63% |
| BrowseComp | — | 69% |
| OSWorld-Verified | — | 58% |
| CodingGemma 4 31B wins | ||
| LiveCodeBench | 80% | 41% |
| HumanEval | — | 66% |
| SWE-bench Verified | — | 42% |
| SWE-bench Pro | — | 47% |
| Multimodal & GroundedGemma 4 31B wins | ||
| MMMU-Pro | 76.9% | 61% |
| OfficeQA Pro | — | 74% |
| ReasoningNemotron 3 Ultra 500B wins | ||
| BBH | 74.4% | 85% |
| MRCRv2 | 66.4% | 85% |
| MuSR | — | 69% |
| LongBench v2 | — | 81% |
| KnowledgeGemma 4 31B wins | ||
| GPQA | 84.3% | 73% |
| MMLU-Pro | 85.2% | 73% |
| HLE | 26.5% | 15% |
| HLE w/o tools | 19.5% | — |
| SuperGPQA | — | 71% |
| FrontierScience | — | 67% |
| Instruction Following | ||
| IFEval | — | 84% |
| Multilingual | ||
| MGSM | — | 81% |
| MMLU-ProX | — | 79% |
| Mathematics | ||
| AIME 2023 | — | 74% |
| HMMT Feb 2023 | — | 70% |
| HMMT Feb 2024 | — | 72% |
| MATH-500 | — | 84% |
Gemma 4 31B is ahead overall, 73 to 60. The biggest single separator in this matchup is LiveCodeBench, where the scores are 80% and 41%.
Gemma 4 31B has the edge for knowledge tasks in this comparison, averaging 61.3 versus 56.1. Inside this category, MMLU-Pro is the benchmark that creates the most daylight between them.
Gemma 4 31B has the edge for coding in this comparison, averaging 80 versus 43.5. Inside this category, LiveCodeBench is the benchmark that creates the most daylight between them.
Nemotron 3 Ultra 500B has the edge for reasoning in this comparison, averaging 79.1 versus 66.4. Inside this category, MRCRv2 is the benchmark that creates the most daylight between them.
Gemma 4 31B has the edge for multimodal and grounded tasks in this comparison, averaging 76.9 versus 66.9. Inside this category, MMMU-Pro 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.