Claude Opus 5 vs Claude Fable 5
Claude Opus 5 (Anthropic) and Claude Fable 5 (Anthropic) compared on price, context window, speed and independent benchmarks. Claude Opus 5 both scores higher on the Artificial Analysis Intelligence Index (50.8 vs 49.6) and is cheaper on a blended basis — 2.0× less than Claude Fable 5.
| Claude Opus 5 | Claude Fable 5 | |
|---|---|---|
| Provider | Anthropic | Anthropic |
| Input $/1M | $5.00 | $10.0 |
| Output $/1M | $25.0 | $50.0 |
| Blended $/1M (3:1) | $10.0 | $20.0 |
| Cached input $/1M | $0.50 | $1.00 |
| Context window | 1M | 1M |
| Knowledge cutoff | 2026-05 | 2026-01 |
| License | proprietary | proprietary |
| Speed (tok/s) | 57 | 68 |
| Artificial Analysis Intelligence Index | 50.8 | 49.6 |
| LMArena Elo | 1505 | 1493 |
| GPQA Diamond | 93.2% | 92.6% |
| AIME | — | 99.7%* |
| SWE-bench Verified | 97% | 95% |
| LiveCodeBench | 74.6% | — |
| MMLU-Pro | 91.6% | 91.5%* |
| Humanity's Last Exam | 54.9% | 55.5% |