Comparar modelos
A metric-by-metric comparison using the latest available benchmark, runtime, and pricing values. Tie wins more comparable metrics.
| Metric | GPT-5.6 Sol (max) | Kimi K3 |
|---|---|---|
| Artificial Analysis Intelligence Index | 58,9 | 57,1 |
| Artificial Analysis Coding Index | 77,4 | 76,2 |
| LiveBench Mathematics | 96,2% | 84,4% |
| GPQA | 94,1% | 93,5% |
| Humanity's Last Exam | 47,2% | 44,3% |
| SciCode | 56,1% | 58,7% |
| Output Speed | 67 tok/s | 38,4 tok/s |
| Time to First Token | 120,38 s | 5,79 s |
| Blended Price | US$ 11,25 / 1M | US$ 6,00 / 1M |
| Input Price | US$ 5,00 / 1M | US$ 3,00 / 1M |
| Output Price |
| US$ 30,00 / 1M |
| US$ 15,00 / 1M |
| Value Index | 5,24 | 9,52 |
