CLAUDE OPUS 5.5

Summary from the TLDR AI newsletter

Anthropic introduced Claude Opus 5.5, saying it matched Claude Fable 5.1 on most work while costing 40% less to run than Opus 5. The model also underwent external evaluations and achieved Anthropic's strongest result to date on its automated behavioral audit.