Compare / Qwen3.8 Max (0902) vs Claude Haiku 4.5
Qwen3.8 Max (0902) vs Claude Haiku 4.5
44 results from 3 sources that measured both models. Each row is in the source's own units; there is no overall winner.
| Benchmark | Metric | Qwen3.8 Max (0902) | Claude Haiku 4.5 |
|---|---|---|---|
| CatalogBenchmeasured by us | Channel rules broken% of products | 3.0% | 22.6% |
| CatalogBenchmeasured by us | Content quality% of checks | 93.3% | 94.2% |
| CatalogBenchmeasured by us | Failed outputs% of products | 3.6% | 0.0% |
| CatalogBenchmeasured by us | Missing UK information% of products | 0.0% | 0.0% |
| CatalogBenchmeasured by us | Not findable% of products | 1.8% | 6.5% |
| CatalogBenchmeasured by us | Publish-ready listings% of products | 10.1% | 1.2% |
| CatalogBenchmeasured by us | Reliably publish-ready% of products | 1.8% | 0.0% |
| CatalogBenchmeasured by us | Unsupported claimsclaims per product | 4.58 | 7.30 |
| CatalogBenchmeasured by us | Unsupported claims% of products | 83.3% | 95.8% |
| CatalogBenchmeasured by us | Wrong attributes% of products | 19.6% | 51.2% |
| CatalogBenchmeasured by us | Wrong category or variant% of products | 2.4% | 6.5% |
| CatalogBenchmeasured by us | Channel compliance% of products | 100.0% | 92.3% |
| CatalogBenchmeasured by us | Channel rules broken% of products | 0.6% | 7.7% |
| CatalogBenchmeasured by us | Conflicts caught% of conflicts | 100.0% | 100.0% |
| CatalogBenchmeasured by us | Content quality% of checks | 94.2% | 94.9% |
| CatalogBenchmeasured by us | Cost per productUS dollars | $0.0256 | $0.0061 |
| CatalogBenchmeasured by us | Decision accuracy% of decisions | 93.4% | 89.3% |
| CatalogBenchmeasured by us | Failed outputs% of products | 3.6% | 0.0% |
| CatalogBenchmeasured by us | Field accuracy% of missing fields | 90.3% | 93.7% |
| CatalogBenchmeasured by us | Invented values% of filled values | 2.7% | 6.2% |
| CatalogBenchmeasured by us | Missing UK information% of products | 0.0% | 0.0% |
| CatalogBenchmeasured by us | Not findable% of products | 3.6% | 8.9% |
| CatalogBenchmeasured by us | Publish-ready listings% of products | 53.0% | 7.7% |
| CatalogBenchmeasured by us | Reliably publish-ready% of products | 33.9% | 1.8% |
| CatalogBenchmeasured by us | Unsupported claims% of products | 22.6% | 80.4% |
| CatalogBenchmeasured by us | Unsupported claimsclaims per product | 0.64 | 3.50 |
| CatalogBenchmeasured by us | Wrong attributes% of products | 21.4% | 48.2% |
| CatalogBenchmeasured by us | Wrong category or variant% of products | 1.8% | 7.7% |
| ROASBenchmeasured by us | Audience scorescore out of 100 | 39.7 | 30.9 |
| ROASBenchmeasured by us | Business scorescore out of 100 | 15.6 | 7.3 |
| ROASBenchmeasured by us | Consistency scorescore out of 100 | 19.2 | 8.7 |
| ROASBenchmeasured by us | Contribution profitsimulated US dollars | $99,775 | −$205,001 |
| ROASBenchmeasured by us | Cost of a runUS dollars | $0.73 | $0.0298 |
| ROASBenchmeasured by us | Months over budgetmonths of 12 | 2 | 2 |
| ROASBenchmeasured by us | Overall scorescore out of 100 | 25.4 | 17.5 |
| ROASBenchmeasured by us | Planning scorescore out of 100 | 54.8 | 54.4 |
| ROASBenchmeasured by us | Return on ad spendprofit per $1 spent | $0.09 | −$0.21 |
| Arena (formerly LMArena)reported | OverallArena rating | 1,471 | 1,418 |
| Arena (formerly LMArena)reported | Business, management and financeArena rating | 1,480 | 1,420 |
| Arena (formerly LMArena)reported | Creative writingArena rating | 1,467 | 1,387 |
| Arena (formerly LMArena)reported | Expert promptsArena rating | 1,510 | 1,452 |
| Arena (formerly LMArena)reported | Instruction followingArena rating | 1,472 | 1,415 |
| Arena (formerly LMArena)reported | OverallArena rating | 1,479 | 1,414 |
| Arena (formerly LMArena)reported | Writing, literature and languageArena rating | 1,472 | 1,399 |
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.