Compare / gemini-2.5-flash-preview-09-2025 vs GPT-5.1
gemini-2.5-flash-preview-09-2025 vs GPT-5.1
17 results from 3 sources that measured both models. Each row is in the source's own units; there is no overall winner.
| Benchmark | Metric | gemini-2.5-flash-preview-09-2025 | GPT-5.1 |
|---|---|---|---|
| Arena (formerly LMArena)reported | OverallArena rating | 1,425 | 1,459 |
| Arena (formerly LMArena)reported | Business, management and financeArena rating | 1,403 | 1,448 |
| Arena (formerly LMArena)reported | Creative writingArena rating | 1,380 | 1,428 |
| Arena (formerly LMArena)reported | Expert promptsArena rating | 1,432 | 1,482 |
| Arena (formerly LMArena)reported | Instruction followingArena rating | 1,400 | 1,450 |
| Arena (formerly LMArena)reported | OverallArena rating | 1,403 | 1,456 |
| Arena (formerly LMArena)reported | Writing, literature and languageArena rating | 1,394 | 1,437 |
| Artificial Analysisreported | Artificial Analysis Intelligence Indexindex score | 15.5 | 24.7 |
| Artificial Analysisreported | GPQA Diamond% of questions | 79.3% | 87.3% |
| Artificial Analysisreported | Humanity's Last Exam% of questions | 13.8% | 28.5% |
| Artificial Analysisreported | IFBench% of instructions | 52.3% | 72.9% |
| Artificial Analysisreported | Long-context reasoning (AA-LCR)% of questions | 71.0% | 80.0% |
| Artificial Analysisreported | Terminal-Bench Hard% of tasks | 16.7% | 45.5% |
| Artificial Analysisreported | Τ²-bench telecom% of tasks | 45.6% | 81.9% |
| UGI Leaderboardreported | Requested-length error% off the requested word count | 20.0% | 113.0% |
| UGI Leaderboardreported | Style adherencescore from 0 to 1 | 0.34 | 0.38 |
| UGI Leaderboardreported | Writing scorescore out of 100 | 44.9 | 55.8 |
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.