Compare / Gemma 4 31B vs GPT-5 Nano
Gemma 4 31B vs GPT-5 Nano
16 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 Nano |
|---|---|---|---|
| Arena (formerly LMArena)reported | Business, management and financeArena rating | 1,449 | 1,339 |
| Arena (formerly LMArena)reported | Creative writingArena rating | 1,419 | 1,247 |
| Arena (formerly LMArena)reported | Expert promptsArena rating | 1,499 | 1,354 |
| Arena (formerly LMArena)reported | Instruction followingArena rating | 1,453 | 1,327 |
| Arena (formerly LMArena)reported | OverallArena rating | 1,453 | 1,338 |
| Arena (formerly LMArena)reported | Writing, literature and languageArena rating | 1,435 | 1,289 |
| Artificial Analysisreported | Artificial Analysis Intelligence Indexindex score | 14.7 | 13.0 |
| Artificial Analysisreported | GPQA Diamond% of questions | 85.7% | 67.6% |
| Artificial Analysisreported | Humanity's Last Exam% of questions | 23.6% | 9.5% |
| Artificial Analysisreported | IFBench% of instructions | 75.6% | 67.6% |
| Artificial Analysisreported | Long-context reasoning (AA-LCR)% of questions | 69.7% | 45.0% |
| Artificial Analysisreported | Terminal-Bench Hard% of tasks | 36.4% | 17.4% |
| Artificial Analysisreported | Τ²-bench telecom% of tasks | 65.5% | 36.5% |
| SimpleQA Verified (Epoch AI)reported | Correct answers% of questions | 10.4% | 11.7% |
| Vectara Hallucination Leaderboardreported | Answer rate% of documents | 100.0% | 100.0% |
| Vectara Hallucination Leaderboardreported | Hallucination rate% of summaries | 7.4% | 10.5% |
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.