Compare / Gemma 4 31B vs GPT-5.6 Terra
Gemma 4 31B vs GPT-5.6 Terra
24 results from 4 sources that measured both models. Each row is in the source's own units; there is no overall winner.
| Benchmark | Metric | Gemma 4 31B | GPT-5.6 Terra |
|---|---|---|---|
| Arena (formerly LMArena)reported | OverallArena rating | 1,445 | 1,477 |
| Arena (formerly LMArena)reported | OverallArena rating | 1,450 | 1,467 |
| Arena (formerly LMArena)reported | Business, management and financeArena rating | 1,449 | 1,466 |
| Arena (formerly LMArena)reported | Creative writingArena rating | 1,419 | 1,428 |
| Arena (formerly LMArena)reported | Expert promptsArena rating | 1,499 | 1,504 |
| Arena (formerly LMArena)reported | Instruction followingArena rating | 1,453 | 1,462 |
| Arena (formerly LMArena)reported | OverallArena rating | 1,453 | 1,465 |
| Arena (formerly LMArena)reported | Writing, literature and languageArena rating | 1,435 | 1,445 |
| Artificial Analysisreported | Artificial Analysis Coding Indexindex score | 43.4 | 76.7 |
| Artificial Analysisreported | Artificial Analysis Intelligence Indexindex score | 14.7 | 42.1 |
| Artificial Analysisreported | GPQA Diamond% of questions | 85.7% | 92.5% |
| Artificial Analysisreported | Humanity's Last Exam% of questions | 23.6% | 42.9% |
| Artificial Analysisreported | IFBench% of instructions | 75.6% | 71.2% |
| Artificial Analysisreported | Long-context reasoning (AA-LCR)% of questions | 69.7% | 83.0% |
| Artificial Analysisreported | SciCode% of problems | 45.5% | 55.0% |
| Artificial Analysisreported | Terminal-Bench 2.1% of tasks | 43.4% | 88.0% |
| Artificial Analysisreported | Terminal-Bench 4.0% of tasks | 0.0% | 35.4% |
| Artificial Analysisreported | Terminal-Bench Hard% of tasks | 36.4% | 62.9% |
| Artificial Analysisreported | Τ-bench banking% of tasks | 14.8% | 40.2% |
| Artificial Analysisreported | Τ²-bench telecom% of tasks | 65.5% | 86.3% |
| SimpleQA Verified (Epoch AI)reported | Correct answers% of questions | 10.4% | 43.2% |
| UGI Leaderboardreported | Requested-length error% off the requested word count | 5.0% | 22.0% |
| UGI Leaderboardreported | Style adherencescore from 0 to 1 | 0.36 | 0.39 |
| UGI Leaderboardreported | Writing scorescore out of 100 | 43.1 | 60.4 |
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.