Anthropic Launches Claude Sonnet 4.5 for Advanced Programming

Anthropic has unveiled its latest AI model, Claude Sonnet 4.5, which boasts advanced capabilities in the field of programming. The company claims that this model can generate 'production-ready' applications.

According to the blog, this is the most consistent frontier model released by Anthropic, with significant improvements over previous versions of Claude. Over the past year, Anthropic's solutions have gained popularity for addressing software development tasks, with companies like Apple and Meta reportedly using Claude models internally.

Claude Sonnet 4.5 demonstrates industry-leading performance in several programming benchmarks, including SWE-Bench Verified. The model can autonomously write code for up to 30 hours, during which it not only created an application but also deployed database services, registered domain names, and conducted audits according to the SOC 2 standard.

Alongside the launch of Claude Sonnet 4.5, Anthropic introduced the Claude Agent SDK, an infrastructure for creating agents. The company demonstrated how this extension allows Claude to operate directly in the browser, navigating websites, filling out spreadsheets, and completing tasks.

Anthropic noted significant progress in defending against prompt injection attacks, where attackers insert hidden instructions into prompts to make the model behave unsafely or break rules. The neural network will be available via API and in the Claude chatbot, with costs set at $3 per million input tokens and $15 for output tokens.

In August, Anthropic programmed the Claude Opus 4 and 4.1 chatbots to end dialogues with users in 'rare, extreme cases of systematically harmful or abusive interactions.'