Anthropic Unveils Claude Opus 5.5, Its Strongest Behavioral-Alignment Model Yet
Summary
Anthropic unveils Claude Opus 5.5 as its strongest behavioral-alignment model, restricting its use for cybersecurity and biology tasks to vetted organizations while pricing it at $4 per million input tokens and $20 per million output tokens.
Key Points
- Anthropic unveils Claude Opus 5.5, calling it its strongest-performing model yet for behavioral alignment and adding safeguards for its most capable model.
- Cybersecurity tasks are generally rerouted to Opus 4.8 and biology-classifier requests to Opus 5; only organizations vetted through Anthropic's Life Sciences and Cyber programs can use Opus 5.5 for those tasks.
- Opus 5.5 costs 40% less than Opus 5 on typical workloads, charges $4 per million input tokens and $20 per million output tokens, and generates outputs more than 30% faster.