Anthropic has announced the release of Claude 3.7 Sonnet, a new hybrid reasoning AI model, alongside a specialized tool for developers called Claude Code. Claude 3.7 Sonnet is described as the company’s most advanced model to date and introduces a unique approach by combining rapid-response capabilities with extended, self-reflective reasoning. This dual-mode functionality allows users to choose between quick answers or more in-depth, step-by-step problem-solving. The model is available across all Anthropic plans—Free, Pro, Team, and Enterprise—as well as through platforms like Amazon Bedrock and Google Cloud’s Vertex AI. However, the extended thinking mode is excluded from the Free tier.
Introducing Claude 3.7 Sonnet: our most intelligent model to date. It's a hybrid reasoning model, producing near-instant responses or extended, step-by-step thinking.
— Anthropic (@AnthropicAI) February 24, 2025
One model, two ways to think.
We’re also releasing an agentic coding tool: Claude Code. pic.twitter.com/jt7qQmFWuC
Key improvements in Claude 3.7 Sonnet include enhanced performance in coding and front-end web development tasks, making it particularly effective for real-world applications such as debugging, planning code changes, and handling complex workflows. The model also introduces fine-grained control over its "thinking budget" via API integration, enabling users to set limits on how much computational effort the model should expend on a task.
Accompanying this release is Claude Code, a command-line tool aimed at streamlining software development workflows. Currently available as a limited research preview, Claude Code empowers developers to automate tasks such as reading and editing code, running tests, and managing GitHub repositories directly from their terminal. Early testing has shown that the tool can significantly reduce development time by completing tasks that would typically take over 45 minutes in just one pass.
Anthropic has emphasized safety and reliability in this release. The system card for Claude 3.7 Sonnet outlines measures taken to mitigate risks like prompt injection attacks while improving the model's ability to distinguish between harmful and benign requests. Compared to its predecessor, it reduces unnecessary refusals by 45%.
This release reflects Anthropic's broader goal of creating AI systems capable of augmenting human capabilities across various domains. By integrating reasoning into a single model and focusing on practical applications rather than theoretical benchmarks, Claude 3.7 Sonnet aims to set a new standard for large language models in both business and development contexts. Feedback from users of both Claude 3.7 Sonnet and Claude Code will play a critical role in shaping future updates and improvements to these tools.
