Compare / Claude Opus 5 vs Gemma 4 31B
Claude Opus 5 vs Gemma 4 31B
18 results from 3 sources that measured both models. Each row is in the source's own units; there is no overall winner.
| Benchmark | Metric | Claude Opus 5 | Gemma 4 31B |
|---|---|---|---|
| Arena (formerly LMArena)reported | OverallArena rating | 1,490 | 1,445 |
| Arena (formerly LMArena)reported | OverallArena rating | 1,488 | 1,450 |
| Arena (formerly LMArena)reported | Business, management and financeArena rating | 1,486 | 1,449 |
| Arena (formerly LMArena)reported | Creative writingArena rating | 1,472 | 1,419 |
| Arena (formerly LMArena)reported | Expert promptsArena rating | 1,543 | 1,499 |
| Arena (formerly LMArena)reported | Instruction followingArena rating | 1,497 | 1,453 |
| Arena (formerly LMArena)reported | OverallArena rating | 1,491 | 1,453 |
| Arena (formerly LMArena)reported | Writing, literature and languageArena rating | 1,482 | 1,435 |
| Artificial Analysisreported | Artificial Analysis Coding Indexindex score | 78.0 | 43.4 |
| Artificial Analysisreported | Artificial Analysis Intelligence Indexindex score | 50.8 | 14.7 |
| Artificial Analysisreported | GPQA Diamond% of questions | 93.7% | 85.7% |
| Artificial Analysisreported | Humanity's Last Exam% of questions | 54.9% | 23.6% |
| Artificial Analysisreported | Long-context reasoning (AA-LCR)% of questions | 82.0% | 69.7% |
| Artificial Analysisreported | SciCode% of problems | 56.4% | 45.5% |
| Artificial Analysisreported | Terminal-Bench 2.1% of tasks | 89.1% | 43.4% |
| Artificial Analysisreported | Terminal-Bench 4.0% of tasks | 49.0% | 0.0% |
| Artificial Analysisreported | Τ-bench banking% of tasks | 44.7% | 14.8% |
| SimpleQA Verified (Epoch AI)reported | Correct answers% of questions | 59.9% | 10.4% |
Bold green marks the better value on that metric. Where a source reports ranges that overlap, the difference may not be meaningful; see the benchmark page for ranges.