Compare / Claude Opus 4.8 vs Kimi K3
Claude Opus 4.8 vs Kimi K3
26 results from 7 sources that measured both models. Each row is in the source's own units; there is no overall winner.
| Benchmark | Metric | Claude Opus 4.8 | Kimi K3 |
|---|---|---|---|
| APEX-Agentsreported | Tasks passed% of tasks | 48.9% | 50.6% |
| Arena (formerly LMArena)reported | Confirmed task successIPS effect estimate | 0.04 | 0.09 |
| Arena (formerly LMArena)reported | Praise over complaintIPS effect estimate | 0.15 | 0.07 |
| Arena (formerly LMArena)reported | SteerabilityIPS effect estimate | 0.10 | -0.01 |
| Arena (formerly LMArena)reported | Tool groundingIPS effect estimate | -0.00 | 0.00 |
| Arena (formerly LMArena)reported | OverallArena rating | 1,469 | 1,472 |
| Arena (formerly LMArena)reported | Business, management and financeArena rating | 1,490 | 1,489 |
| Arena (formerly LMArena)reported | Creative writingArena rating | 1,467 | 1,458 |
| Arena (formerly LMArena)reported | Expert promptsArena rating | 1,525 | 1,533 |
| Arena (formerly LMArena)reported | Instruction followingArena rating | 1,491 | 1,486 |
| Arena (formerly LMArena)reported | OverallArena rating | 1,480 | 1,488 |
| Arena (formerly LMArena)reported | Writing, literature and languageArena rating | 1,474 | 1,475 |
| Artificial Analysisreported | Artificial Analysis Coding Indexindex score | 74.3 | 76.2 |
| Artificial Analysisreported | Artificial Analysis Intelligence Indexindex score | 41.8 | 43.6 |
| Artificial Analysisreported | GPQA Diamond% of questions | 92.0% | 93.5% |
| Artificial Analysisreported | Humanity's Last Exam% of questions | 48.7% | 46.9% |
| Artificial Analysisreported | Long-context reasoning (AA-LCR)% of questions | 77.7% | 88.7% |
| Artificial Analysisreported | SciCode% of problems | 54.4% | 59.5% |
| Artificial Analysisreported | Terminal-Bench 2.1% of tasks | 84.6% | 85.0% |
| Artificial Analysisreported | Terminal-Bench 4.0% of tasks | 21.7% | 12.6% |
| Artificial Analysisreported | Τ-bench banking% of tasks | 34.2% | 46.0% |
| GDP.pdfreported | Professional document tasks% of rubric | 24.0% | 19.0% |
| SimpleQA Verified (Epoch AI)reported | Correct answers% of questions | 53.0% | 50.6% |
| tau2-benchreported | Consistency (pass^4)% of tasks | 22.7% | 17.5% |
| tau2-benchreported | Task success (pass^1)% of tasks | 39.7% | 37.1% |
| Vending-Bench 2reported | Money after a yearUS dollars | $5,787.43 | $5,165.03 |
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.