Anthropic has released Claude Opus 5.5, the first model in its new Claude 5.5 family. In its September 22 announcement, the company said the model performs at the level of Claude Fable 5.1 on most work while costing 40% less to run than Claude Opus 5.
The listed API price is $4 per million input tokens and $20 per million output tokens, versus $5 and $25 for Opus 5. Anthropic also cut its listed cache-read price from $0.50 to $0.20 per million tokens, a meaningful line item for agents and coding tools that repeatedly reuse long context. The company says Opus 5.5 generates output more than 30% faster than its predecessor.
Company claims, and what they mean
Anthropic says Frontier Design and METR tested Opus 5.5 before release. It also says the model produced its strongest result so far on the company's automated behavioral audit and is more resistant than Opus 5 to prompt injection. Those results, like the benchmark comparisons in the launch material, are company-reported evaluations rather than independent performance verdicts.
For developers, the more concrete change is the pricing and the model's intended role: Anthropic is positioning Opus 5.5 for complex coding, computer use and long-running agent work. The actual cost and performance will still depend on a workload's prompts, tools, effort settings and cache use, but the launch makes the top Opus tier cheaper on the company’s published rate card.
