Compare / gemini-3-pro vs GPT-5.4
gemini-3-pro vs GPT-5.4
15 results from 2 sources that measured both models. Each row is in the source's own units; there is no overall winner.
| Benchmark | Metric | gemini-3-pro | GPT-5.4 |
|---|---|---|---|
| Arena (formerly LMArena)reported | OverallArena rating | 1,451 | 1,470 |
| Arena (formerly LMArena)reported | OverallArena rating | 1,201 | 1,195 |
| Arena (formerly LMArena)reported | OverallArena rating | 1,481 | 1,485 |
| Arena (formerly LMArena)reported | Business, management and financeArena rating | 1,474 | 1,482 |
| Arena (formerly LMArena)reported | Creative writingArena rating | 1,484 | 1,443 |
| Arena (formerly LMArena)reported | Expert promptsArena rating | 1,498 | 1,515 |
| Arena (formerly LMArena)reported | Instruction followingArena rating | 1,473 | 1,471 |
| Arena (formerly LMArena)reported | OverallArena rating | 1,485 | 1,475 |
| Arena (formerly LMArena)reported | Writing, literature and languageArena rating | 1,481 | 1,465 |
| OpenHands Indexreported | Average% resolved | 49.0% | 64.3% |
| OpenHands Indexreported | Front end (SWE-Bench Multimodal)% resolved | 36.8% | 36.8% |
| OpenHands Indexreported | Greenfield (Commit0)% resolved | 25.0% | 56.2% |
| OpenHands Indexreported | Information gathering (GAIA)% resolved | 44.2% | 82.4% |
| OpenHands Indexreported | Issue resolution (SWE-Bench)% resolved | 70.6% | 75.6% |
| OpenHands Indexreported | Testing (SWT-Bench)% resolved | 68.6% | 70.4% |
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.