Anthropic has launched Claude Opus 5.5, the first model in the Claude 5.5 series. This model achieves performance comparable to Fable 5.1 in most tasks, reduces costs for typical workloads by 40% compared to Opus 5, and increases output speed by over 30%. Meanwhile, prices for API calls and cache reads have also been significantly reduced. In coding evaluations, Claude Opus 5.5 excels, scoring higher than GPT-6 Astra in multiple assessments while offering significantly lower costs per question. However, it still falls short of Astra in projects like AutomationBench and Terminal-Bench-Science. Notably, this is the first model released after Anthropic announced a slowdown in advancing frontier models, and it currently achieves the best automated behavior audit scores. The five-hour usage quota for relevant paying users has been increased accordingly, and Anthropic plans to release two additional models, Sonnet 5.5 and Haiku 5.5, in the future.