Side-by-side benchmark comparison across agentic, coding, multimodal, knowledge, reasoning, and math workflows.
Gemma 4 31B
73
2/8 categoriesGLM-4.7
74
Winner · 2/8 categoriesGemma 4 31B· GLM-4.7
Pick GLM-4.7 if you want the stronger benchmark profile. Gemma 4 31B only becomes the better choice if coding is the priority or you need the larger 256K context window.
GLM-4.7 finishes one point ahead overall, 74 to 73. That is enough to call, but not enough to treat as a blowout. This matchup comes down to a few meaningful edges rather than one model dominating the board.
GLM-4.7's sharpest advantage is in knowledge, where it averages 74.8 against 61.3. The single biggest benchmark swing on the page is MRCRv2, 66.4% to 78%. Gemma 4 31B does hit back in coding, so the answer changes if that is the part of the workload you care about most.
Gemma 4 31B gives you the larger context window at 256K, compared with 200K for GLM-4.7.
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 | GLM-4.7 |
|---|---|---|
| Agentic | ||
| OSWorld-Verified | — | 61% |
| VITA-Bench | — | 15.5% |
| CodingGemma 4 31B wins | ||
| LiveCodeBench | 80% | 84.9% |
| HumanEval | — | 94.2% |
| SWE-bench Verified | — | 73.8% |
| SWE-bench Pro | — | 51% |
| Multimodal & GroundedGemma 4 31B wins | ||
| MMMU-Pro | 76.9% | 66% |
| OfficeQA Pro | — | 76% |
| ReasoningGLM-4.7 wins | ||
| BBH | 74.4% | 84% |
| MRCRv2 | 66.4% | 78% |
| MuSR | — | 80% |
| LongBench v2 | — | 79% |
| KnowledgeGLM-4.7 wins | ||
| GPQA | 84.3% | 85.7% |
| MMLU-Pro | 85.2% | 84.3% |
| HLE | 26.5% | — |
| HLE w/o tools | 19.5% | — |
| MMLU | — | 86% |
| SuperGPQA | — | 82% |
| FrontierScience | — | 72% |
| SimpleQA | — | 46% |
| Instruction Following | ||
| Coming soon | ||
| Multilingual | ||
| MGSM | — | 94% |
| MMLU-ProX | — | 78% |
| Mathematics | ||
| AIME 2023 | — | 86% |
| AIME 2024 | — | 88% |
| AIME 2025 | — | 95.7% |
| HMMT Feb 2023 | — | 82% |
| HMMT Feb 2025 | — | 97.1% |
| BRUMO 2025 | — | 85% |
GLM-4.7 is ahead overall, 74 to 73. The biggest single separator in this matchup is MRCRv2, where the scores are 66.4% and 78%.
GLM-4.7 has the edge for knowledge tasks in this comparison, averaging 74.8 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 69.3. Inside this category, LiveCodeBench is the benchmark that creates the most daylight between them.
GLM-4.7 has the edge for reasoning in this comparison, averaging 78.9 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 70.5. 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.