Compare / Gemini 3.7 Flash vs GPT-5.6 Luna
Gemini 3.7 Flash vs GPT-5.6 Luna
25 results from 5 sources that measured both models. Each row is in the source's own units; there is no overall winner.
| Benchmark | Metric | Gemini 3.7 Flash | GPT-5.6 Luna |
|---|---|---|---|
| Arena (formerly LMArena)reported | Confirmed task successIPS effect estimate | 0.00 | -0.05 |
| Arena (formerly LMArena)reported | Praise over complaintIPS effect estimate | -0.03 | -0.04 |
| Arena (formerly LMArena)reported | SteerabilityIPS effect estimate | -0.00 | 0.02 |
| Arena (formerly LMArena)reported | Tool groundingIPS effect estimate | -0.00 | 0.00 |
| Arena (formerly LMArena)reported | OverallArena rating | 1,477 | 1,461 |
| Arena (formerly LMArena)reported | Business, management and financeArena rating | 1,483 | 1,463 |
| Arena (formerly LMArena)reported | Creative writingArena rating | 1,492 | 1,412 |
| Arena (formerly LMArena)reported | Expert promptsArena rating | 1,512 | 1,491 |
| Arena (formerly LMArena)reported | Instruction followingArena rating | 1,487 | 1,446 |
| Arena (formerly LMArena)reported | OverallArena rating | 1,488 | 1,454 |
| Arena (formerly LMArena)reported | Writing, literature and languageArena rating | 1,496 | 1,435 |
| Artificial Analysisreported | Artificial Analysis Coding Indexindex score | 76.1 | 71.4 |
| Artificial Analysisreported | Artificial Analysis Intelligence Indexindex score | 39.6 | 37.3 |
| Artificial Analysisreported | GPQA Diamond% of questions | 94.5% | 91.1% |
| Artificial Analysisreported | Humanity's Last Exam% of questions | 47.9% | 39.5% |
| Artificial Analysisreported | Long-context reasoning (AA-LCR)% of questions | 83.0% | 83.7% |
| Artificial Analysisreported | SciCode% of problems | 59.8% | 53.6% |
| Artificial Analysisreported | Terminal-Bench 2.1% of tasks | 85.8% | 80.9% |
| Artificial Analysisreported | Terminal-Bench 4.0% of tasks | 13.6% | 11.6% |
| Artificial Analysisreported | Τ-bench banking% of tasks | 35.5% | 31.1% |
| GDP.pdfreported | Professional document tasks% of rubric | 23.8% | 22.7% |
| SimpleQA Verified (Epoch AI)reported | Correct answers% of questions | 69.2% | 41.0% |
| UGI Leaderboardreported | Requested-length error% off the requested word count | 7.0% | 15.0% |
| UGI Leaderboardreported | Style adherencescore from 0 to 1 | 0.33 | 0.38 |
| UGI Leaderboardreported | Writing scorescore out of 100 | 77.5 | 53.7 |
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.