New model targets software development and advanced AI workloads while Anthropic expands its safety testing.

SAN FRANCISCO — September 23, 2026 — Anthropic has launched Claude Opus 5.5, a new artificial intelligence model designed to improve performance while reducing operating costs as competition among major AI developers intensifies.

According to Reuters, Anthropic says Claude Opus 5.5 delivers performance comparable to its previous high-end systems while costing about 40% less to operate. The company also said the model performed strongly on software-development benchmarks.

The new model arrives amid growing attention on AI safety and the rapid development of advanced AI systems. Anthropic said Opus 5.5 underwent external testing by AI safety research groups before its release and includes safeguards used in some of its more capable models.

Anthropic has priced Claude Opus 5.5 at $4 per million input tokens and $20 per million output tokens, according to Reuters. The model is available through major cloud platforms including Amazon Web Services, Google Cloud and Microsoft Azure.

The company also plans to release Claude Sonnet 5.5 and Claude Haiku 5.5 in the coming weeks, expanding its latest model family.

The launch adds another major development to the rapidly changing U.S. AI industry, where companies are competing on model performance, cost, safety and enterprise adoption.