- Published on
Morning Report...
- Authors
- Name
- DeepHorus
Anthropic has recently released its latest artificial intelligence model, Claude Fable 5.1, marking another breakthrough in AI technology in terms of cost-effectiveness and performance. According to the company's disclosure, Fable 5.1, through optimized cache read pricing, reduces costs by approximately 25% compared to its predecessor Fable 5 for typical token-based tasks. For highly agentic and complex tasks, this cost saving can reach up to 45%. The new model has demonstrated excellent performance in multiple benchmark tests, scoring 55.8% in Terminal-Bench 4.0, higher than Fable 5's 42.0%; and 73.4% in CursorBench 3.2.0, also surpassing the previous generation's 70.5%.\n\nThis update is not a single version release. Anthropic simultaneously launched Claude Mythos 5.1, which is based on the same underlying model but is provided through a trusted access program, with protection mechanisms specifically tailored for sensitive areas such as cybersecurity and life sciences. The cybersecurity capabilities of Fable 5.1 have also been enhanced, reducing the false interception rate of normal requests by 60% and allowing users to search for software vulnerabilities in a controlled environment, while strictly blocking the development of exploit code.\n\nFor enterprise users, Anthropic's enterprise-level service, Enterprise Frontier Safeguards, emphasizes data sovereignty, keeping customer data within cloud infrastructure controlled by the customers themselves. This service is scheduled to be rolled out in phases starting this autumn, aiming to meet the higher demands of enterprises for data security and compliance.\nThe release of Fable 5.1 is a significant step in the AI model race, directly improving cost-effectiveness and potentially prompting other AI service providers to follow suit in optimizing their cost structures. Its enhanced security features also reflect the industry's increasing emphasis on the responsible development of AI technology, especially in critical application areas such as cybersecurity.\nLooking ahead, with the continuous improvement of model capabilities and ongoing cost optimization, AI technology is expected to be deeply applied in more vertical industries. Through this update, Anthropic has not only solidified its technological position in the field of large language models but also paved the way for the popularization of enterprise-level AI, which is anticipated to accelerate intelligent processes in areas such as office work, research, and security.
