On October 15, 2025, Anthropic launched Claude Haiku 4.5, the latest iteration of its scaled-down AI model. This version is touted to deliver performance akin to Sonnet 4 but at a significantly reduced cost and increased speed. In internal benchmarks, Haiku 4.5 achieved a score of 73% on SWE-Bench and 41% on Terminal-Bench, aligning it with other models like GPT-5 and Gemini 2.5. The lightweight nature of Haiku 4.5 allows for the deployment of multiple agents simultaneously, making it particularly suitable for production environments. Anthropic's CPO, Mike Krieger, emphasized that this model opens new possibilities for AI deployment, especially in software development where speed is crucial. The introduction of Haiku 4.5 follows recent launches of other models by Anthropic, indicating the company's rapid innovation in the AI space.