Side-by-side benchmark comparison across agentic, coding, multimodal, knowledge, reasoning, and math workflows.
Gemini 3 Pro Deep Think
80
Winner · 3/8 categoriesGemma 4 31B
73
1/8 categoriesGemini 3 Pro Deep Think· Gemma 4 31B
Pick Gemini 3 Pro Deep Think if you want the stronger benchmark profile. Gemma 4 31B only becomes the better choice if coding is the priority.
Gemini 3 Pro Deep Think is clearly ahead on the aggregate, 80 to 73. The gap is large enough that you do not need to squint at the spreadsheet to see the difference.
Gemini 3 Pro Deep Think's sharpest advantage is in multimodal & grounded, where it averages 95 against 76.9. The single biggest benchmark swing on the page is MRCRv2, 96% to 66.4%. Gemma 4 31B does hit back in coding, so the answer changes if that is the part of the workload you care about most.
Gemini 3 Pro Deep Think gives you the larger context window at 2M, 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 | Gemini 3 Pro Deep Think | Gemma 4 31B |
|---|---|---|
| Agentic | ||
| Terminal-Bench 2.0 | 77% | — |
| BrowseComp | 87% | — |
| OSWorld-Verified | 73% | — |
| CodingGemma 4 31B wins | ||
| HumanEval | 91% | — |
| SWE-bench Verified | 58% | — |
| LiveCodeBench | 58% | 80% |
| Multimodal & GroundedGemini 3 Pro Deep Think wins | ||
| MMMU-Pro | 95% | 76.9% |
| OfficeQA Pro | 95% | — |
| ReasoningGemini 3 Pro Deep Think wins | ||
| MuSR | 93% | — |
| BBH | 95% | 74.4% |
| LongBench v2 | 94% | — |
| MRCRv2 | 96% | 66.4% |
| ARC-AGI-2 | 45.1% | — |
| KnowledgeGemini 3 Pro Deep Think wins | ||
| MMLU | 99% | — |
| GPQA | 97% | 84.3% |
| SuperGPQA | 95% | — |
| MMLU-Pro | 81% | 85.2% |
| HLE | 32% | 26.5% |
| FrontierScience | 88% | — |
| SimpleQA | 95% | — |
| HLE w/o tools | — | 19.5% |
| Instruction Following | ||
| IFEval | 89% | — |
| Multilingual | ||
| MGSM | 92% | — |
| MMLU-ProX | 85% | — |
| Mathematics | ||
| AIME 2023 | 99% | — |
| AIME 2024 | 99% | — |
| HMMT Feb 2024 | 97% | — |
| HMMT Feb 2025 | 96% | — |
| BRUMO 2025 | 96% | — |
| MATH-500 | 92% | — |
Gemini 3 Pro Deep Think is ahead overall, 80 to 73. The biggest single separator in this matchup is MRCRv2, where the scores are 96% and 66.4%.
Gemini 3 Pro Deep Think has the edge for knowledge tasks in this comparison, averaging 76.4 versus 61.3. 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 58. Inside this category, LiveCodeBench is the benchmark that creates the most daylight between them.
Gemini 3 Pro Deep Think has the edge for reasoning in this comparison, averaging 82.1 versus 66.4. Inside this category, MRCRv2 is the benchmark that creates the most daylight between them.
Gemini 3 Pro Deep Think has the edge for multimodal and grounded tasks in this comparison, averaging 95 versus 76.9. 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.