Anthropic Launches Claude Opus 5.5, Citing Lower Running Costs
Anthropic has released its latest flagship AI model, Claude Opus 5.5, which the company states offers improved performance and significantly lower operational costs compared to its predecessor.
AI company Anthropic has launched its newest flagship model, Claude Opus 5.5. According to the company, the model delivers enhanced performance, faster output, and strengthened safeguards, while also reducing running costs significantly.
Anthropic reports that Claude Opus 5.5 performs at the level of its more powerful predecessor, Claude Fable 5.1, on most tasks. The company claims it costs approximately 40% less to run on typical workloads than the previous Opus 5 model and generates output over 30% faster. While the token pricing has decreased by 20%, the estimated 40% cost reduction for users stems from the new model's efficiency in using fewer tokens to complete tasks.
The model is being positioned for agentic coding, computer use, and knowledge work. This release comes amidst increasing competition in the AI sector, with rivals like OpenAI also introducing new models. Anthropic's offering aims to provide a more cost-effective and efficient solution for demanding AI applications.
Safety features have also been enhanced in Opus 5.5. Anthropic highlighted the model's improved ability to avoid attempts to bypass containment boundaries during safety evaluations. The company has previously advocated for pacing the development of frontier AI to allow safety practices and policies to evolve accordingly.
Claude Opus 5.5 is available to subscribers of various Claude plans and accessible to developers through major cloud platforms like Amazon Web Services, Google Cloud, and Microsoft Azure. Updates, Claude Sonnet 5.5 and Haiku 5.5, are expected soon with similar improvements.