Modelle vergleichen
A metric-by-metric comparison using the latest available benchmark, runtime, and pricing values. Kimi K3 (max) wins more comparable metrics.
| Metric | Claude Opus 5 (Adaptive Reasoning, Max Effort) | Kimi K3 (max) |
|---|---|---|
| Artificial Analysis Intelligence Index | 60,7 | 57,1 |
| Artificial Analysis Coding Index | 78 | 76,2 |
| LiveBench Mathematics | 95,7% | 84,4% |
| GPQA | 93,2% | 93,5% |
| Humanity's Last Exam | 52,6% | 44,3% |
| SciCode | 55,7% | 58,7% |
| Output Speed | 59,1 tok/s | 34,5 tok/s |
| Time to First Token | 34,11 s | 2,94 s |
| Blended Price | 10,00 $ / 1M | 6,00 $ / 1M |
| Input Price | 5,00 $ / 1M | 3,00 $ / 1M |
| Output Price | 25,00 $ / 1M | 15,00 $ / 1M |
| Value Index | 6,07 | 9,52 |
