Comparar modelos
A metric-by-metric comparison using the latest available benchmark, runtime, and pricing values. Claude Opus 5 (Adaptive Reasoning, Max Effort) wins more comparable metrics.
| Metric | Claude Opus 5 (Adaptive Reasoning, Max Effort) | GPT-5.6 Sol (max) |
|---|---|---|
| Artificial Analysis Intelligence Index | 60,7 | 58,9 |
| Artificial Analysis Coding Index | 78 | 77,4 |
| LiveBench Mathematics | 95,7% | 96,2% |
| GPQA | 93,2% | 94,1% |
| Humanity's Last Exam | 52,6% | 47,2% |
| SciCode | 55,7% | 56,1% |
| Output Speed | 43,9 tok/s | 73,9 tok/s |
| Time to First Token | 28,7 s | 86,49 s |
| Blended Price | US$ 10,00 / 1M | US$ 11,25 / 1M |
| Input Price | US$ 5,00 / 1M | US$ 5,00 / 1M |
| Output Price | US$ 25,00 / 1M | US$ 30,00 / 1M |
| Value Index | 6,07 | 5,24 |
