Claude Opus 5.5 just dropped – first in the 5.5 lineup.
Performance-wise, it matches Claude Fable 5.1 across most benchmarks, but runs 40% cheaper than Opus 5. That's a solid cost-performance trade-off if you're running high-volume inference workloads.
Basically: same capability tier as Fable 5.1, but with Opus 5's pricing slashed. Worth testing if you're currently on Opus 5 and want to cut API costs without downgrading output quality.
Performance-wise, it matches Claude Fable 5.1 across most benchmarks, but runs 40% cheaper than Opus 5. That's a solid cost-performance trade-off if you're running high-volume inference workloads.
Basically: same capability tier as Fable 5.1, but with Opus 5's pricing slashed. Worth testing if you're currently on Opus 5 and want to cut API costs without downgrading output quality.