Compare / Claude Fable 5.1 vs DeepSeek V4.1 Flash
Claude Fable 5.1 vs DeepSeek V4.1 Flash
46 results from 3 sources that measured both models. Each row is in the source's own units; there is no overall winner.
| Benchmark | Metric | Claude Fable 5.1 | DeepSeek V4.1 Flash |
|---|---|---|---|
| CatalogBenchmeasured by us | Channel rules broken% of products | 0.0% | 1.8% |
| CatalogBenchmeasured by us | Claims to check% of products | 0.0% | 0.0% |
| CatalogBenchmeasured by us | Content quality% of checks | 97.2% | 96.2% |
| CatalogBenchmeasured by us | Failed outputs% of products | 0.0% | 0.6% |
| CatalogBenchmeasured by us | Missing UK information% of products | 0.0% | 0.0% |
| CatalogBenchmeasured by us | Not findable% of products | 1.8% | 3.6% |
| CatalogBenchmeasured by us | Publish-ready listings% of products | 13.7% | 47.0% |
| CatalogBenchmeasured by us | Reliably publish-ready% of products | 5.4% | 23.2% |
| CatalogBenchmeasured by us | Unsupported claims% of products | 83.9% | 39.9% |
| CatalogBenchmeasured by us | Unsupported claimsclaims per product | 3.86 | 1.25 |
| CatalogBenchmeasured by us | Wrong attributes% of products | 14.9% | 20.2% |
| CatalogBenchmeasured by us | Wrong category or variant% of products | 1.8% | 2.4% |
| CatalogBenchmeasured by us | Channel compliance% of products | 100.0% | 99.4% |
| CatalogBenchmeasured by us | Channel rules broken% of products | 0.0% | 3.0% |
| CatalogBenchmeasured by us | Claims to check% of products | 0.6% | 0.0% |
| CatalogBenchmeasured by us | Conflicts caught% of conflicts | 100.0% | 100.0% |
| CatalogBenchmeasured by us | Content quality% of checks | 97.6% | 96.7% |
| CatalogBenchmeasured by us | Cost per productUS dollars | $0.0991 | $0.0061 |
| CatalogBenchmeasured by us | Decision accuracy% of decisions | 95.9% | 96.9% |
| CatalogBenchmeasured by us | Failed outputs% of products | 0.0% | 0.0% |
| CatalogBenchmeasured by us | Field accuracy% of missing fields | 94.7% | 94.4% |
| CatalogBenchmeasured by us | Invented values% of filled values | 0.7% | 3.1% |
| CatalogBenchmeasured by us | Missing UK information% of products | 0.0% | 0.0% |
| CatalogBenchmeasured by us | Not findable% of products | 1.8% | 3.6% |
| CatalogBenchmeasured by us | Publish-ready listings% of products | 46.4% | 63.7% |
| CatalogBenchmeasured by us | Reliably publish-ready% of products | 26.8% | 41.1% |
| CatalogBenchmeasured by us | Unsupported claimsclaims per product | 1.23 | 0.54 |
| CatalogBenchmeasured by us | Unsupported claims% of products | 48.8% | 13.1% |
| CatalogBenchmeasured by us | Wrong attributes% of products | 11.9% | 20.2% |
| CatalogBenchmeasured by us | Wrong category or variant% of products | 2.4% | 3.0% |
| Arena (formerly LMArena)reported | Confirmed task successIPS effect estimate | 0.17 | 0.12 |
| Arena (formerly LMArena)reported | Praise over complaintIPS effect estimate | 0.32 | 0.03 |
| Arena (formerly LMArena)reported | SteerabilityIPS effect estimate | 0.08 | -0.03 |
| Arena (formerly LMArena)reported | Tool groundingIPS effect estimate | 0.00 | 0.00 |
| Arena (formerly LMArena)reported | OverallArena rating | 1,501 | 1,471 |
| Arena (formerly LMArena)reported | Business, management and financeArena rating | 1,485 | 1,481 |
| Arena (formerly LMArena)reported | Creative writingArena rating | 1,484 | 1,440 |
| Arena (formerly LMArena)reported | Expert promptsArena rating | 1,516 | 1,518 |
| Arena (formerly LMArena)reported | Instruction followingArena rating | 1,497 | 1,478 |
| Arena (formerly LMArena)reported | OverallArena rating | 1,501 | 1,477 |
| Arena (formerly LMArena)reported | Writing, literature and languageArena rating | 1,496 | 1,455 |
| Artificial Analysisreported | Artificial Analysis Intelligence Indexindex score | 53.4 | 39.5 |
| Artificial Analysisreported | Humanity's Last Exam% of questions | 59.1% | 39.2% |
| Artificial Analysisreported | Long-context reasoning (AA-LCR)% of questions | 85.3% | 84.0% |
| Artificial Analysisreported | SciCode% of problems | 63.1% | 51.9% |
| Artificial Analysisreported | Terminal-Bench 4.0% of tasks | 55.1% | 26.8% |
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.