Side-by-side benchmark comparison across agentic, coding, multimodal, knowledge, reasoning, and math workflows.
Gemma 4 E2B
~39
Winner · 3/8 categoriesMistral 7B v0.3
29
1/8 categoriesGemma 4 E2B· Mistral 7B v0.3
Pick Gemma 4 E2B if you want the stronger benchmark profile. Mistral 7B v0.3 only becomes the better choice if reasoning is the priority or you would rather avoid the extra latency and token burn of a reasoning model.
Gemma 4 E2B is clearly ahead on the aggregate, 39 to 29. The gap is large enough that you do not need to squint at the spreadsheet to see the difference.
Gemma 4 E2B's sharpest advantage is in coding, where it averages 44 against 13.5. The single biggest benchmark swing on the page is BBH, 21.9% to 63%. Mistral 7B v0.3 does hit back in reasoning, so the answer changes if that is the part of the workload you care about most.
Gemma 4 E2B is the reasoning model in the pair, while Mistral 7B v0.3 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. Gemma 4 E2B gives you the larger context window at 128K, compared with 32K for Mistral 7B v0.3.
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 E2B | Mistral 7B v0.3 |
|---|---|---|
| Agentic | ||
| Terminal-Bench 2.0 | — | 24% |
| BrowseComp | — | 32% |
| OSWorld-Verified | — | 25% |
| CodingGemma 4 E2B wins | ||
| LiveCodeBench | 44% | 14% |
| HumanEval | — | 30.5% |
| SWE-bench Verified | — | 15% |
| SWE-bench Pro | — | 12% |
| Multimodal & GroundedGemma 4 E2B wins | ||
| MMMU-Pro | 44.2% | 27% |
| OfficeQA Pro | — | 39% |
| ReasoningMistral 7B v0.3 wins | ||
| BBH | 21.9% | 63% |
| MRCRv2 | 19.1% | 41% |
| MuSR | — | 26% |
| LongBench v2 | — | 38% |
| KnowledgeGemma 4 E2B wins | ||
| GPQA | 43.4% | 29% |
| MMLU-Pro | 60% | 54% |
| SuperGPQA | — | 27% |
| HLE | — | 5% |
| FrontierScience | — | 34% |
| SimpleQA | — | 28% |
| Instruction Following | ||
| IFEval | — | 68% |
| Multilingual | ||
| MGSM | — | 62% |
| MMLU-ProX | — | 60% |
| Mathematics | ||
| AIME 2023 | — | 30% |
| AIME 2024 | — | 32% |
| AIME 2025 | — | 31% |
| HMMT Feb 2023 | — | 26% |
| HMMT Feb 2024 | — | 28% |
| HMMT Feb 2025 | — | 27% |
| BRUMO 2025 | — | 29% |
| MATH-500 | — | 60% |
Gemma 4 E2B is ahead overall, 39 to 29. The biggest single separator in this matchup is BBH, where the scores are 21.9% and 63%.
Gemma 4 E2B has the edge for knowledge tasks in this comparison, averaging 54.1 versus 29.5. Inside this category, GPQA is the benchmark that creates the most daylight between them.
Gemma 4 E2B has the edge for coding in this comparison, averaging 44 versus 13.5. Inside this category, LiveCodeBench is the benchmark that creates the most daylight between them.
Mistral 7B v0.3 has the edge for reasoning in this comparison, averaging 35.8 versus 19.1. Inside this category, BBH is the benchmark that creates the most daylight between them.
Gemma 4 E2B has the edge for multimodal and grounded tasks in this comparison, averaging 44.2 versus 32.4. 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.