Compare / GPT-5.6 Sol vs GPT-6 Luna
GPT-5.6 Sol vs GPT-6 Luna
18 results from 4 sources that measured both models. Each row is in the source's own units; there is no overall winner.
| Benchmark | Metric | GPT-5.6 Sol | GPT-6 Luna |
|---|---|---|---|
| Arena (formerly LMArena)reported | Confirmed task successIPS effect estimate | 0.03 | 0.05 |
| Arena (formerly LMArena)reported | Praise over complaintIPS effect estimate | 0.19 | -0.03 |
| Arena (formerly LMArena)reported | SteerabilityIPS effect estimate | 0.07 | 0.02 |
| Arena (formerly LMArena)reported | Tool groundingIPS effect estimate | 0.00 | 0.00 |
| Arena (formerly LMArena)reported | OverallArena rating | 1,476 | 1,422 |
| Arena (formerly LMArena)reported | Business, management and financeArena rating | 1,485 | 1,452 |
| Arena (formerly LMArena)reported | Creative writingArena rating | 1,470 | 1,403 |
| Arena (formerly LMArena)reported | Expert promptsArena rating | 1,534 | 1,505 |
| Arena (formerly LMArena)reported | Instruction followingArena rating | 1,489 | 1,443 |
| Arena (formerly LMArena)reported | OverallArena rating | 1,483 | 1,442 |
| Arena (formerly LMArena)reported | Writing, literature and languageArena rating | 1,483 | 1,415 |
| Artificial Analysisreported | Artificial Analysis Intelligence Indexindex score | 47.0 | 38.1 |
| Artificial Analysisreported | Humanity's Last Exam% of questions | 49.5% | 38.5% |
| Artificial Analysisreported | Long-context reasoning (AA-LCR)% of questions | 84.0% | 83.3% |
| Artificial Analysisreported | SciCode% of problems | 57.8% | 54.6% |
| Artificial Analysisreported | Terminal-Bench 4.0% of tasks | 39.9% | 12.6% |
| GDP.pdfreported | Professional document tasks% of rubric | 30.7% | 23.0% |
| SimpleQA Verified (Epoch AI)reported | Correct answers% of questions | 69.7% | 41.4% |
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.