Compare / Claude Fable 5 vs MiniMax M3
Claude Fable 5 vs MiniMax M3
20 results from 4 sources that measured both models. Each row is in the source's own units; there is no overall winner.
| Benchmark | Metric | Claude Fable 5 | MiniMax M3 |
|---|---|---|---|
| APEX-Agentsreported | Tasks passed% of tasks | 63.6% | 37.7% |
| Arena (formerly LMArena)reported | Confirmed task successIPS effect estimate | 0.03 | -0.14 |
| Arena (formerly LMArena)reported | Praise over complaintIPS effect estimate | 0.17 | -0.13 |
| Arena (formerly LMArena)reported | SteerabilityIPS effect estimate | 0.11 | -0.07 |
| Arena (formerly LMArena)reported | Tool groundingIPS effect estimate | 0.00 | -0.00 |
| Arena (formerly LMArena)reported | OverallArena rating | 1,499 | 1,433 |
| Arena (formerly LMArena)reported | OverallArena rating | 1,483 | 1,434 |
| Arena (formerly LMArena)reported | Business, management and financeArena rating | 1,503 | 1,450 |
| Arena (formerly LMArena)reported | Creative writingArena rating | 1,501 | 1,406 |
| Arena (formerly LMArena)reported | Expert promptsArena rating | 1,550 | 1,471 |
| Arena (formerly LMArena)reported | Instruction followingArena rating | 1,509 | 1,435 |
| Arena (formerly LMArena)reported | OverallArena rating | 1,504 | 1,440 |
| Arena (formerly LMArena)reported | Writing, literature and languageArena rating | 1,508 | 1,422 |
| OpenHands Indexreported | Average% resolved | 81.0% | 57.2% |
| OpenHands Indexreported | Front end (SWE-Bench Multimodal)% resolved | 70.6% | 36.8% |
| OpenHands Indexreported | Greenfield (Commit0)% resolved | 62.5% | 25.0% |
| OpenHands Indexreported | Information gathering (GAIA)% resolved | 84.2% | 66.7% |
| OpenHands Indexreported | Issue resolution (SWE-Bench)% resolved | 95.8% | 76.4% |
| OpenHands Indexreported | Testing (SWT-Bench)% resolved | 91.9% | 81.1% |
| Vending-Bench 2reported | Money after a yearUS dollars | $5,680.26 | $2,157.77 |
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.