Side-by-side benchmark comparison across agentic, coding, multimodal, knowledge, reasoning, and math workflows.
Gemma 4 26B A4B
64
Winner · 4/8 categoriesMistral 8x7B v0.2
27
0/8 categoriesGemma 4 26B A4B· Mistral 8x7B v0.2
Pick Gemma 4 26B A4B if you want the stronger benchmark profile. Mistral 8x7B v0.2 only becomes the better choice if you would rather avoid the extra latency and token burn of a reasoning model.
Gemma 4 26B A4B is clearly ahead on the aggregate, 64 to 27. The gap is large enough that you do not need to squint at the spreadsheet to see the difference.
Gemma 4 26B A4B's sharpest advantage is in coding, where it averages 77.1 against 12. The single biggest benchmark swing on the page is LiveCodeBench, 77.1% to 12%.
Gemma 4 26B A4B is the reasoning model in the pair, while Mistral 8x7B v0.2 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 26B A4B gives you the larger context window at 256K, compared with 32K for Mistral 8x7B v0.2.
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 26B A4B | Mistral 8x7B v0.2 |
|---|---|---|
| Agentic | ||
| Terminal-Bench 2.0 | — | 24% |
| BrowseComp | — | 34% |
| CodingGemma 4 26B A4B wins | ||
| LiveCodeBench | 77.1% | 12% |
| Multimodal & GroundedGemma 4 26B A4B wins | ||
| MMMU-Pro | 73.8% | 26% |
| OfficeQA Pro | — | 40% |
| ReasoningGemma 4 26B A4B wins | ||
| BBH | 64.8% | 62% |
| MRCRv2 | 44.1% | — |
| MuSR | — | 25% |
| LongBench v2 | — | 39% |
| KnowledgeGemma 4 26B A4B wins | ||
| GPQA | 82.3% | 28% |
| MMLU-Pro | 82.6% | — |
| HLE | 17.2% | — |
| HLE w/o tools | 8.7% | — |
| MMLU | — | 29% |
| SuperGPQA | — | 26% |
| SimpleQA | — | 27% |
| Instruction Following | ||
| IFEval | — | 67% |
| Multilingual | ||
| MGSM | — | 62% |
| Mathematics | ||
| AIME 2025 | — | 30% |
| HMMT Feb 2025 | — | 26% |
Gemma 4 26B A4B is ahead overall, 64 to 27. The biggest single separator in this matchup is LiveCodeBench, where the scores are 77.1% and 12%.
Gemma 4 26B A4B has the edge for knowledge tasks in this comparison, averaging 56.1 versus 27. Inside this category, GPQA is the benchmark that creates the most daylight between them.
Gemma 4 26B A4B has the edge for coding in this comparison, averaging 77.1 versus 12. Inside this category, LiveCodeBench is the benchmark that creates the most daylight between them.
Gemma 4 26B A4B has the edge for reasoning in this comparison, averaging 44.1 versus 33.4. Inside this category, BBH is the benchmark that creates the most daylight between them.
Gemma 4 26B A4B has the edge for multimodal and grounded tasks in this comparison, averaging 73.8 versus 32.3. 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.