Compare Models
A metric-by-metric comparison using the latest available benchmark, runtime, and pricing values. Tie wins more comparable metrics.
| Metric | Claude Opus 5 (Adaptive Reasoning, Max Effort) | Grok 4.5 (high) |
|---|---|---|
| Artificial Analysis Intelligence Index | 60.7 | 53.8 |
| Artificial Analysis Coding Index | 78 | 72.4 |
| LiveBench Mathematics | 95.7% | 90.8% |
| GPQA | 93.2% | 93.1% |
| Humanity's Last Exam | 52.6% | 40.3% |
| SciCode | 55.7% | 54.1% |
| Output Speed | 43.9 tok/s | 55.8 tok/s |
| Time to First Token | 28.7 s | 12.63 s |
| Blended Price | $10.00 / 1M | $3.00 / 1M |
| Input Price | $5.00 / 1M | $2.00 / 1M |
| Output Price | $25.00 / 1M | $6.00 / 1M |
| Value Index | 6.07 | 17.93 |
