Claude Opus 5.5 vs Claude Opus 5
Claude Opus 5.5 (Anthropic) and Claude Opus 5 (Anthropic) compared on price, context window, speed and independent benchmarks. Claude Opus 5.5 both scores higher on the Artificial Analysis Intelligence Index (57.6 vs 50.8) and is cheaper on a blended basis — 1.3× less than Claude Opus 5.
| Claude Opus 5.5 | Claude Opus 5 | |
|---|---|---|
| Provider | Anthropic | Anthropic |
| Input $/1M | $4.00 | $5.00 |
| Output $/1M | $20.0 | $25.0 |
| Blended $/1M (3:1) | $8.00 | $10.0 |
| Cached input $/1M | $0.20 | $0.50 |
| Context window | 1M | 1M |
| Knowledge cutoff | 2026-06 | 2026-05 |
| License | proprietary | proprietary |
| Speed (tok/s) | 85 | 57 |
| Artificial Analysis Intelligence Index | 57.6 | 50.8 |
| LMArena Elo | — | 1505 |
| GPQA Diamond | — | 93.2% |
| SWE-bench Verified | — | 97% |
| LiveCodeBench | — | 74.6% |
| MMLU-Pro | — | 91.6% |
| Humanity's Last Exam | 61.4% | 54.9% |