Side-by-side benchmark comparison across agentic, coding, multimodal, knowledge, reasoning, and math workflows.
Gemma 4 31B
73
Winner · 4/8 categoriesNemotron Ultra 253B
41
0/8 categoriesGemma 4 31B· Nemotron Ultra 253B
Pick Gemma 4 31B if you want the stronger benchmark profile. Nemotron Ultra 253B only becomes the better choice if its workflow or ecosystem matters more than the raw scoreboard.
Gemma 4 31B is clearly ahead on the aggregate, 73 to 41. 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 33.3. The single biggest benchmark swing on the page is LiveCodeBench, 80% to 30%.
Gemma 4 31B gives you the larger context window at 256K, compared with 32K for Nemotron Ultra 253B.
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 Ultra 253B |
|---|---|---|
| Agentic | ||
| Terminal-Bench 2.0 | — | 46% |
| BrowseComp | — | 50% |
| OSWorld-Verified | — | 45% |
| CodingGemma 4 31B wins | ||
| LiveCodeBench | 80% | 30% |
| HumanEval | — | 41% |
| SWE-bench Verified | — | 31% |
| SWE-bench Pro | — | 38% |
| Multimodal & GroundedGemma 4 31B wins | ||
| MMMU-Pro | 76.9% | 37% |
| OfficeQA Pro | — | 54% |
| ReasoningGemma 4 31B wins | ||
| BBH | 74.4% | 77% |
| MRCRv2 | 66.4% | 56% |
| MuSR | — | 45% |
| LongBench v2 | — | 55% |
| KnowledgeGemma 4 31B wins | ||
| GPQA | 84.3% | 48% |
| MMLU-Pro | 85.2% | 63% |
| HLE | 26.5% | 11% |
| HLE w/o tools | 19.5% | — |
| MMLU | — | 49% |
| SuperGPQA | — | 46% |
| FrontierScience | — | 49% |
| SimpleQA | — | 47% |
| Instruction Following | ||
| IFEval | — | 78% |
| Multilingual | ||
| MGSM | — | 74% |
| MMLU-ProX | — | 68% |
| Mathematics | ||
| AIME 2023 | — | 49% |
| AIME 2024 | — | 51% |
| AIME 2025 | — | 50% |
| HMMT Feb 2023 | — | 45% |
| HMMT Feb 2024 | — | 47% |
| HMMT Feb 2025 | — | 46% |
| BRUMO 2025 | — | 48% |
| MATH-500 | — | 74% |
Gemma 4 31B is ahead overall, 73 to 41. The biggest single separator in this matchup is LiveCodeBench, where the scores are 80% and 30%.
Gemma 4 31B has the edge for knowledge tasks in this comparison, averaging 61.3 versus 42.6. Inside this category, GPQA is the benchmark that creates the most daylight between them.
Gemma 4 31B has the edge for coding in this comparison, averaging 80 versus 33.3. Inside this category, LiveCodeBench is the benchmark that creates the most daylight between them.
Gemma 4 31B has the edge for reasoning in this comparison, averaging 66.4 versus 52.7. 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 44.7. 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.