| Claude Fable 5.2 | Claude Opus 5.5 | |
|---|---|---|
| CTO Index | 95.5 (#1) | 95.3 (#2) |
| Provider | Anthropic | Anthropic |
| Weights | Closed | Closed |
| Price $/M in / out | $8 / $40 | $4 / $20 |
| Context window | 1M | 1M |
| License | Proprietary | Proprietary |
| Terminal-Bench 4.1 | 65.8% | 65.8% |
| WebDev Arena | 1824 | 1819 |
| ARC-AGI-2 | 93.8% | 93.3% |
| Text Arena | 1503 | 1506 |
| ARC-AGI-3 | 81.2% | — |
Snapshot 22 November 2026 · Scores copied from each benchmark's own leaderboard · How the CTO Index works
Frequently asked questions
Is Claude Fable 5.2 better than Claude Opus 5.5?
Claude Fable 5.2 scores higher on the CTO Index (95.5 vs 95.3). Across the 4 benchmarks both have results, Claude Fable 5.2 wins 2 (WebDev, ARC-AGI-2) and Claude Opus 5.5 wins 1 (Text Arena). Claude Opus 5.5 costs 50% less per blended token.
Which is better for coding agents, Claude Fable 5.2 or Claude Opus 5.5?
On Terminal-Bench 4.1, Claude Fable 5.2 scores 65.8% against Claude Opus 5.5's 65.8%. Differences under about three points are within the error bars.
How do Claude Fable 5.2 and Claude Opus 5.5 compare on price?
Claude Fable 5.2 lists at $8 / $40 and Claude Opus 5.5 at $4 / $20 per million input / output tokens.