Compare / Gemini 3.8 Flash vs GPT-6 Luna
Gemini 3.8 Flash vs GPT-6 Luna
58 results from 4 sources that measured both models. Each row is in the source's own units; there is no overall winner.
| Benchmark | Metric | Gemini 3.8 Flash | GPT-6 Luna |
|---|---|---|---|
| BulletBenchmeasured by us | Cost per gameUS dollars | $0.0795 | $0.0072 |
| BulletBenchmeasured by us | Games lost on time% of games | 12.5% | 50.0% |
| BulletBenchmeasured by us | Invalid moves% of moves | 0.0% | 0.0% |
| BulletBenchmeasured by us | Ladder Eloladder Elo | 1,126 | 557 |
| BulletBenchmeasured by us | Median move timemilliseconds | 4.0 s | 7.0 s |
| BulletBenchmeasured by us | Cost per gameUS dollars | $0.0154 | $0.0016 |
| BulletBenchmeasured by us | Games lost on time% of games | 50.0% | 50.0% |
| BulletBenchmeasured by us | Invalid moves% of moves | 0.0% | 0.0% |
| BulletBenchmeasured by us | Ladder Eloladder Elo | 1,068 | 249 |
| BulletBenchmeasured by us | Median move timemilliseconds | 1.8 s | 1.5 s |
| CatalogBenchmeasured by us | Channel rules broken% of products | 3.0% | 0.0% |
| CatalogBenchmeasured by us | Content quality% of checks | 97.3% | 97.6% |
| CatalogBenchmeasured by us | Failed outputs% of products | 0.0% | 0.0% |
| CatalogBenchmeasured by us | Missing UK information% of products | 0.0% | 0.0% |
| CatalogBenchmeasured by us | Not findable% of products | 8.9% | 10.1% |
| CatalogBenchmeasured by us | Publish-ready listings% of products | 2.4% | 61.9% |
| CatalogBenchmeasured by us | Reliably publish-ready% of products | 1.8% | 42.9% |
| CatalogBenchmeasured by us | Unsupported claimsclaims per product | 7.12 | 0.56 |
| CatalogBenchmeasured by us | Unsupported claims% of products | 97.0% | 16.7% |
| CatalogBenchmeasured by us | Wrong attributes% of products | 11.9% | 19.1% |
| CatalogBenchmeasured by us | Wrong category or variant% of products | 3.0% | 0.0% |
| CatalogBenchmeasured by us | Channel compliance% of products | 98.8% | 100.0% |
| CatalogBenchmeasured by us | Channel rules broken% of products | 1.2% | 0.0% |
| CatalogBenchmeasured by us | Conflicts caught% of conflicts | 100.0% | 100.0% |
| CatalogBenchmeasured by us | Content quality% of checks | 97.1% | 97.2% |
| CatalogBenchmeasured by us | Cost per productUS dollars | $0.0103 | $0.0005 |
| CatalogBenchmeasured by us | Decision accuracy% of decisions | 94.3% | 96.7% |
| CatalogBenchmeasured by us | Failed outputs% of products | 0.0% | 0.0% |
| CatalogBenchmeasured by us | Field accuracy% of missing fields | 92.7% | 93.7% |
| CatalogBenchmeasured by us | Invented values% of filled values | 0.2% | 1.7% |
| CatalogBenchmeasured by us | Missing UK information% of products | 0.0% | 0.0% |
| CatalogBenchmeasured by us | Not findable% of products | 8.3% | 10.1% |
| CatalogBenchmeasured by us | Publish-ready listings% of products | 33.3% | 64.3% |
| CatalogBenchmeasured by us | Reliably publish-ready% of products | 16.1% | 51.8% |
| CatalogBenchmeasured by us | Unsupported claimsclaims per product | 1.79 | 0.41 |
| CatalogBenchmeasured by us | Unsupported claims% of products | 60.1% | 12.5% |
| CatalogBenchmeasured by us | Wrong attributes% of products | 15.5% | 18.4% |
| CatalogBenchmeasured by us | Wrong category or variant% of products | 3.6% | 1.8% |
| ROASBenchmeasured by us | Audience scorescore out of 100 | 67.1 | 68.2 |
| ROASBenchmeasured by us | Business scorescore out of 100 | 51.0 | 54.8 |
| ROASBenchmeasured by us | Consistency scorescore out of 100 | 36.4 | 35.8 |
| ROASBenchmeasured by us | Contribution profitsimulated US dollars | $842,432 | $738,878 |
| ROASBenchmeasured by us | Cost of a runUS dollars | $0.16 | $0.0142 |
| ROASBenchmeasured by us | Months over budgetmonths of 12 | 0 | 0 |
| ROASBenchmeasured by us | Overall scorescore out of 100 | 51.4 | 53.4 |
| ROASBenchmeasured by us | Planning scorescore out of 100 | 54.7 | 54.8 |
| ROASBenchmeasured by us | Return on ad spendprofit per $1 spent | $0.65 | $0.66 |
| Arena (formerly LMArena)reported | Confirmed task successIPS effect estimate | 0.09 | 0.05 |
| Arena (formerly LMArena)reported | Praise over complaintIPS effect estimate | 0.05 | -0.03 |
| Arena (formerly LMArena)reported | SteerabilityIPS effect estimate | 0.04 | 0.02 |
| Arena (formerly LMArena)reported | Tool groundingIPS effect estimate | 0.00 | 0.00 |
| Arena (formerly LMArena)reported | OverallArena rating | 1,477 | 1,422 |
| Arena (formerly LMArena)reported | Business, management and financeArena rating | 1,482 | 1,452 |
| Arena (formerly LMArena)reported | Creative writingArena rating | 1,482 | 1,403 |
| Arena (formerly LMArena)reported | Expert promptsArena rating | 1,529 | 1,505 |
| Arena (formerly LMArena)reported | Instruction followingArena rating | 1,488 | 1,443 |
| Arena (formerly LMArena)reported | OverallArena rating | 1,492 | 1,442 |
| Arena (formerly LMArena)reported | Writing, literature and languageArena rating | 1,486 | 1,415 |
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.