Anthropic Launches Claude Opus 5.5 With Lower Operating Costs
Anthropic launches Claude Opus 5.5 with lower costs and enhanced AI safety measures.
Anthropic on Tuesday launched Claude Opus 5.5, a new artificial intelligence model that the company says delivers performance comparable to its top-tier Fable 5.1 while costing 40 per cent less to operate than its predecessor.
The launch comes as concerns over AI safety continue to grow. Anthropic CEO Dario Amodei earlier this month called on the global AI industry to slow the release of new capabilities while researchers address safety risks.
According to Anthropic, independent AI safety groups Frontier Design and METR tested Opus 5.5 before its release. The company also said the model includes safeguards previously reserved for its most capable systems.
Opus 5.5 Targets Software Development
Anthropic said Opus 5.5 outperformed OpenAI’s GPT-5.6 Sol on a software development benchmark.
The company also claims that the new model costs roughly one-third as much to run on that benchmark.
Pricing provides another difference. Anthropic set the cost at $4 per million input tokens and $20 per million output tokens.
Pakistan Hockey and Kabaddi Teams Maintain Strong Asian Games Starts
The company said those rates are 20pc lower than the previous Opus 5 model.
Model Available Across Major Cloud Platforms
Users can access Opus 5.5 through several major cloud platforms.
The model is available through Amazon Web Services, Google Cloud and Microsoft Azure.
Anthropic is also expanding access for selected researchers. Vetted cybersecurity and life-sciences researchers will receive broader access through the company’s research programmes.
Anthropic Highlights Safety Improvements
Safety remains a major focus of the new release.
Anthropic said Opus 5.5 performed better than earlier systems in its internal safety evaluations.
The company also reported a significant reduction in attempts to bypass containment boundaries. In a dedicated evaluation, Opus 5.5 was about 85pc less likely to make such attempts than Opus 5 or Mythos 5.1.
Anthropic said the model underwent external testing before launch. The company described the testing as part of its effort to evaluate potential risks associated with more capable AI systems.
More Claude Models Coming Soon
Anthropic plans to expand the Claude 5.5 lineup in the coming weeks.
The company said Claude Sonnet 5.5 and Claude Haiku 5.5 will follow.
Both models will bring many of the same performance, speed and safety improvements, according to Anthropic.
The latest launch comes as AI companies face increasing pressure to balance rapid model development with safety research and safeguards.
