Claude Fable 5.2 vs Claude Opus 5.5

As of November 9, 2026, Claude Fable 5.2 costs $8.00 input and $40.00 output per 1M tokens, and Claude Opus 5.5 costs $4.00 and $20.00. Claude Opus 5.5 is the cheaper of the two.

This is the question every Anthropic customer faces after November 9, 2026, and Anthropic's own answer is not to switch by default. Claude Opus 5.5 remains the recommended starting point for most workloads, and Claude Fable 5.2 is the opt-in top tier above it. The rate card makes the trade explicit: Fable 5.2 lists $8 per million input tokens and $40 output, exactly twice the $4 and $20 for Opus 5.5, although cache reads land at the same $0.20 per million on both, because Fable 5.2 bills cache hits at 0.025x base input and Opus 5.5 at 0.05x. A workload dominated by a large cached prefix therefore pays a much smaller premium than the headline rate suggests. Both carry a 1 million token context window as default and maximum with 128K output. What the extra money buys shows up on Anthropic's launch table: FrontierCode 2.0 at 49.1 percent at xhigh against 45.5 for Opus 5.5, and Humanity's Last Exam with tools at 74.2 against 67.7. Fable 5.2 also posts 41.3 percent on the Remote Labor Index at high effort, against 17.9 for Fable 5.1, on a row with no Opus 5.5 figure. The independent picture is just starting: Vals AI puts Fable 5.2 first of 36 models on its Vals Index at 71.31 percent as of November 9, 2026, where Opus 5.5 scored 66.97 on the September 30 board. The split is not total. On AA-Briefcase, Fable 5.2 leads the rubric score 66.1 to 61.2 while Opus 5.5 holds the higher Elo, 1800 against 1785. For latency, Fable 5.2 adds a fast mode at $16 and $80 that generates up to 2.5x faster. All launch-table figures are vendor-reported.