Anthropic has released Claude Opus 5.5, a new version of its flagship AI model that brings significant improvements across multiple dimensions. The model excels at large-scale code migration tasks, software debugging, and generating complete applications from a single command prompt.
Security and reliability have been strengthened considerably. Opus 5.5 is more resistant to prompt injection attacks, less likely to take irreversible actions, and better at respecting defined boundaries—critical improvements for production environments where AI decisions carry real consequences.
Performance gains make the model more practical for widespread adoption. It runs 30% faster than Opus 5 while costing 40% less to operate, with pricing set at $4 per million input tokens and $20 per million output tokens. Opus 5.5 is available now to Pro, Max, and Enterprise subscribers.