Anthropic unveils AI that works 7 hours straight—no breaks needed
What's the story
Anthropic has unveiled its most advanced AI models yet: Claude Opus 4 and Sonnet 4. The models are aimed at setting a new benchmark for AI agents. Anthropic touts Claude Opus 4 as the "best coding model in the world," capable of working autonomously for nearly seven hours, without additional prompts. Both Claude Opus 4 and Sonnet can leverage the web to perform tasks for users, switching between reasoning and tool use.
Scenario
Designed to automate more tedious parts of workday
Anthropic's new model signals a potential shift in workplace AI use, from completing isolated tasks to handling broader goals. According to the company, Claude Opus 4 is designed to automate the more tedious parts of the workday rather than eliminate jobs entirely. "It's the kind of work that might make up 30% of your day, tasks that aren't especially fulfilling or career-building, but are essential for getting your job done," explained Scott White, product lead for Anthropic's AI assistant, Claude.
Information
Claude Sonnet 4 is a significant upgrade to 3.7
Anthropic says Claude Sonnet 4 is a major improvement over Sonnet 3.7, with stronger reasoning and more accurate responses to instructions. Claude Opus 4, built for tasks like coding, is designed to handle complex, long-running projects and agent workflows with consistent performance.
Benchmarks
Take a look at Opus 4's benchmark scores
Opus 4 is Anthropic's most advanced model yet and currently the top coding model in the world, scoring 72.5% on SWE-bench and 43.2% on Terminal-bench. It can maintain high performance for hours on tasks involving thousands of steps. While Sonnet 4 doesn't match Opus 4 in most areas, it's built to strike a balance between power and efficiency, offering a practical option for many everyday uses.
Financial growth
Anthropic's revenue and market position
Anthropic's annualized revenue hit $2 billion in Q1, more than double from the last time. The number of customers spending over $100,000 annually with Anthropic grew eightfold from last year. To increase its liquidity amid intensifying competition in the AI space, the company raised a revolving credit line worth $2.5 billion for five years from Wall Street last week.
Safety measures
Anthropic's approach to safety and user interaction
Anthropic has added stricter safeguards for Opus 4, such as improved harmful content detectors and cybersecurity defenses. The company says internal testing found Opus 4 could "substantially increase" the ability of someone with a STEM background to obtain, produce, or deploy chemical, biological, or nuclear weapons. Both Opus 4 and Sonnet 4 are "hybrid" models capable of near-instant responses as well as extended thinking for deeper reasoning.