Introducing Claude Opus 5.5

Anthropic released Claude Opus 5.5, the first model in its Claude 5.5 family, delivering performance comparable to Claude Fable 5.1 at 40% lower compute cost than Opus 5. Pricing drops to $4 input and $20 output per million tokens, with cache reads at $0.20 per million — a 60% reduction from Opus 5 — and output generation is more than 30% faster. On agentic coding benchmarks, Opus 5.5 leads competitors including GPT-6 Astra and GPT-5.6 Sol on Terminal-Bench 4.0 (66.4%), FrontierCode, and CursorBench, with real-world tests showing a 680,000-line code migration completed in under a day versus weeks for an engineering team. The model also records the highest scores to date on Anthropic's automated behavioral audit, with improved resistance to prompt injection and reduced likelihood of taking hard-to-reverse actions. Claude Sonnet 5.5 and Haiku 5.5 are expected in coming weeks, and Opus 5.5 deploys with safeguards equivalent to Claude Fable 5.1 given comparable capability in biology and cybersecurity domains.